PRODUCTION MCI-MINES AND WEATHER FACTORS
ON MAPLE SAP AN) SUGAR YIELDS IN A
CENTRAL MICHIGAN WOODLOT '

Than” the Degree '0'} Ph. D, '
MICHIGAN STATE UNIVERSITY
. Bingham Maren: C001

.1957 . A

x ; Rizal-419%

THESIS

  

(A):
—3=
N;
to:
00;
fig
0—,—

N=‘

01;

NE

(0%

(NE

(og-

OLE

\ a.) _
If”

This is to certifg that the

thesis entitled
An Investigation of the Effect of Some Production
Techniques and Weather Factors on Maple Sap and
Sugar Yields in a Central Michigan Woodlot

 

presented bg

Bingham M. Cool

has been accepted towards fulfillment

of the requirements for

_Bh.11._ degree in _Eor_e_st1y_ A

it"): n24 ,MT‘“

Major professor

 

Date—lulLlQ.._1L

0-169

5:

 

 

#40"! was

E 1‘ :Lii‘fiab»2’

A egg
W53"

'J

W " ' ' .-.._t

  

APR_ 2 5 ma

, M f,
‘5 .wf
h “6’ 7 a W

 

 

 

ABSTRACT

AN INVESTIGATION OF THE EFFECT OF SOME PRODUCTION TECHNIQUES
AND WEATHER FACTORS 0N MAPLE SAP AND SUGAR YIEDDS
IN A CENTRAL MICHIGAN WOODLOT

A study of some of the aspects of maple sap production
was undertaken at Michigan State University. The experimen-
tal area was approximately six acres of a sixty-eight acre
woodlot on the campus. A total of 168 buckets was hung on
120 trees each year for four years. Daily sap weights and
sugar per cent (total Brix) were recorded for each taphole.
Sap and sugar yields were analyzed statistically for various
treatments as follows:

Twelve trees had taps at one, two, and three feet.
The one-foot tapholes were occasionally covered with snow
and remained frozen, producing significantly lower yields.
No significant differences were found between two- and
three-foot taps.

Twenty-four trees were tested each year with random
pairs of four Spout designs. No significant differences in
yield of sap or sugar were found.

Forty-eight one-bucket trees were used each year to
test tapping depths of two, four, and six inches. Somewhat
inadequate results have been interpreted as indicating that
any gain from increased tapping depth over two inches would
be too slight to offset the additional labor and damage to

the trees.

Thirty-six one-bucket trees were used each year to
test taphole diameters of three-eighths, seven-sixteenths,
eleven-sixteenths, and fifteen-sixteenths inches. Differ-
ences in yields due to taphole diameter were not statis-
tically significant.

For some of the preceding experiments plastic bags
were substituted for metal buckets to catch the sap. Each
year the yields from.those trees with bags and those with
buckets showing no complicating features were analyzed for
differences due to type of container. No significant dif-
ferences were demonstrated. However, each year the yield
from.the plastic bags exceeded that from.the buckets.

All of the experiments had tapholes randomized over
the four cardinal compass quadrants. Any that showed no
obvious bias due to other factors were combined to test dif-
ferences in yields due to compass position. For one year of
the four the taps in the north quadrant produced highly sig-
nificantly less sap than the other three sides. The differ-
ence was not sufficient to justify avoiding the north side
at the risk of concentrating taphole damage in the other
quadrants.

Eighteen trees of approximately seventeen inches in
diameter were included in these experiments--nine having two
tapholes and the other nine only one. Similarly, sixteen

trees were found that were approximately twenty-one inches in

diameter, with eight having three buckets and eight having
only two. The yields from.these trees were analyzed for
four years for differences due to additional tapholes in
trees of a given size. The differences were not statisti-
cally significant.

Daily temperature and humidity data were gathered on
the experimental area. These were supplemented by other
weather observations obtained from the Weather Bureau. An
attempt was made to relate these factors to daily sap flow
and total yield. The relationship between maximum daily
temperature and sap flow was very close. Examination of the
rest of the weather data failed to disclose any significant

trends or correlations.

 

 

 

- 1L1
Approved: 4/; kré’égéb C2 éév/

T‘. D. Stevens
Major Professor

 

AN INVESTIGATION OF THE EFFECT OF SOME PRODUCTION TECHNIQUES
AND WEATHER FACTORS ON MAPLE SAP AND SUGAR YIELDS
IN A CENTRAL MICHIGAN WOODLOT

By

Bingham Mercur Cool

A THESIS
Submitted to the School for Advanced Graduate Studies of
Michigan State University of Agriculture and
Applied Science in partial fulfillment of

the requirements for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry

1957

ACKNOWLEDGEMENTS

The author wishes to express his appreciation to the
following men whose invaluable assistance made this work
possible:

To Dr. T. D. Stevens who was responsible for the ad-
ministration of the research project and, as chairman of the
guidance committee, gave freely of his time and experience
in guiding the graduate program.

i To Professor P. W. Robbins who, as project leader, was
responsible for the planning and supervision of the field work.

To the Agricultural Research Service of the United
States Department of Agriculture, Philadelphia, Pennsylvania,
which, under Cooperative Research Project Number A-ls-33h59,
provided the funds which made this project possible.

To Dr. W. D. Baten for his assistance with the statis-
tical design and analysis.

To Jerry Williams, graduate research assistant, who
selected and classified the trees and recorded the data for
the 1953 and the l95h maple sap seasons.

To Dr. L. M. Turk, Dr. W. B. Drew, and Dr. L. M. James
for their contributions as members of the author's guidance

committee.

TABLE OF CONTENTS

CHAPTER

I.

II.

III.

IV.

VI.

INTRODUCTION . . . . . . . . . . . . . . . .
LOCATION AND DESCRIPTION OF THE EXPERIMENTAL
AREA . . . . . . . . . . . . . . . . . . .
PROCEDURE . . . . . . . . . . . . . . . . .
Application of Treatments . . . . . . . .
Experimental Design . . . . . . . . . . .
HEIGHT OF TAPHOLE FROM GROUND . . . . . . .
Review of Literature . . . . . . . . . . .
Experimental Procedure . . . . . . . . . .
Results and Discussion . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . .
SPOUT DESIGN . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . .
Experimental Procedure . . . . . . . . . .
Results and Discussion . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . .
DEPTH OF TAPHOLE . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . .
Review of Literature . . . . . . . . . ...
Experimental Procedure . . . . . . . . . .
Results and Discussion . . . . . . . . . .

Conclusions . . . . . . . . . . . . . . .

PAGE

11
11L
11L
15
17
20
21
21
22
2h
26
27
27
28
30
31
31+

CHAPTER

VII.

VIII.

IX.

X.

XI.

DIAMETER OF TAPHOLE . .
Introduction . . . .
Experimental Procedure
Results and Discussion
Conclusions . . . . .

TYPE OF CONTAINER . . .
Introduction . . . .
Experimental Procedure
Results and Discussion

Conclusions . . . . .

COMPASS POSITION OF TAPHOLE

Introduction . . . .
Review of Literature
Experimental Procedure
Results and Discussion

Conclusions . . . . .

NUMBER OF TAPHOLES PER TREE

Introduction . . . .
Review of Literature
Experimental Procedure
Results and Discussion
Conclusions . . . .
WEATHER AND SAP FLOW .

Introduction . . . .

iv
PAGE
36
36
38
Al
LB
15
45
50
52
52
55
55
56
58
59
63
61L
6h
65
66
67
68
70
7O

CHAPTER PAGE
Literature Review . . . . . . . . . . . . . . . 71
Experimental Procedure . . . . . . . . . . . . . 73
Results and Discussion . . . . . . . . . . . . . 77
Conclusions . . . . . . . . . . . . . . . . . . 87

XII. CONCLUSIONS AND RECOMMENDATIONS . . . . . . . . . 88
Conclusions . . . . . . . . . . . . . . . . . . 88
Recommendations . . . . . . . . . . . . . . . . 89

LITERATURE CITED . . . . . . . . . . . . . . . . . . . . 93

APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . 98

Analysis I-—Height of Taphole and Compass Position . . 98
Analysis II--Spout Design and Compass Position
(h Years Combined) . . . . . . . . . . . . . . . . . 99
Analysis III-~Depth of Taphole and Compass Position
1956 Season . . . . . . . . . . . . . . . . . . . . 103
Analysis IV--Taphole Diameter and Compass Position
(1953 & 1955 Combined) . . . . . . . . . . . . . . . 106
Analysis V--Taphole Diameter and Compass Position
(195A & 1956 Combined) . . . . . . . . . . . . . . . 109
.Analysis VI--Type of Container (1956 Season) . . . . . 112
'.Ana1ysis VII-~Compass Position (Combined Data from
3 Experiments) 1956 Season . . . . . . . . . . . . . 116
.Analysis VIII-~Number of Buckets Per Tree (Four

Years Data Combined) . . . . . . . . . . . . . . . . 118

LIST OF FIGURES

FIGURE
1. Baker Woodlot Type Map . . . . . . . . . .

2. Recording Rain Gages Adapted for Sap Collecting . .
3. View of Experimental Area Showing Weather Shelter
and Rain Gages . . . . . . . . . . . . .

4. Yield of Maple Sugar by Height of Taphole .
5. Spout Designs Compared for Yield . . . . .
6. Yield of Sap by Spout Design . . . . . . .
7. Yield in Pounds of Sap by Depth of Taphole
8. Oversize Spouts Turned from Aluminum.Rod (A
and Standard Spout (C) . . . . . . . . .
9. Diameter of Taphole and Maple Sap Yield . .
10. Plastic Sap Bag . . . . . . . . . . . . . .
11. Metal Sap Bucket with Snap-on Cover . . . .
12. Emptying Sap Bag into Collecting Pail . . .
13. Mean Yield of Maple Sap by Type of Container
1h. Mean Yield of Maple Sap by Compass Position
15. 1953 Maple Sap Yields and Weather Data . .
16. 1955 Maple Sap Yields and Weather Data . .
17- 1953 Sap Production Compared to Maximum-Minimum
Temperature and Wind . . . . . . . . . .
18. 195h Sugar Production Compared to Maximum-Minimum

Temperature and Wind . . . . . . . . . .

PAGE

12

12
18
23
25
32

NO

LLB
R6
R6
5 3
6O

75

76

78

79

vii

FIGURE PAGE
19. 1955 Sugar Production Compared to Maximum—Minimum

Temperature and Wind . . . . . . . . . . . . . . 8O
20. 1956 Daily Production Compared to Maximum and

Minimum.Temperatures . . . . . . . . . . . . . . 81
21. Total Winter Snowfall Compared to Sap Production . 82
22. Rainfall in Previous Growing Season Compared to

Sap Production . . . . . . . . . . . . . . . . . 83
23. Total Precipitation for Previous Year Compared

to Sap Production . . . . . . . . . . . . . . . . 8h
24. Length of Previous Growing Season Compared to

Sap Production . . . . . . . . . . . . . . . . . 85

CHAPTER I
INTRODUCTION

The maple sirup and sugar industry is not one of the
important cogs in our national economy. However, it is of
considerable importance to a segment of our population with
a relatively low income. It affords an opportunity to sell
farm labor at a time when other farm.operations are not so
pressing. The shortage of sugar during World War II appar-
ently stimulated a renewed interest in maple sugar and sirup,
and, despite the fact that they are now strictly luxury items,
sales are holding up very well.

Alvord (1862) quotes the census of 1850 and of 1860
as reporting maple sugar production in twenty-eight states
east of the Rocky Mountains. However, eleven states from
New England west to Minnesota and south to Kentucky are now
listed as contributing most of the maple sap products in the
'United States. The U. S. Department of Agriculture (1955)
reports an average annual production from.these states be-
tween 19h3 and 195h of 258,000 pounds of maple sugar and
l.76h,000 gallons of sirup. ‘With sugar bringing an average
of slightly over $0.73 per pound and sirup $h.08 per gallon,
this production represents an annual cash income of over
$7,300,000. Michigan has ranked fourth or fifth among these

states in sirup production during this period with an average

annual yield of 96,000 gallons of sirup and 8,333 pounds of
sugar. At the average prices quoted above, this represents
an annual income for the state of almost $h00,000.

The development of the maple sap products industry has
followed more nearly the pattern of an art than a science.
Most early writers and historians credit the American Indian
with having discovered the use of maple sap for sweetening
and with having developed crude methods of concentrating it.
Others, however, point out that the beginnings are so ob-
scure that there is room for doubt as to whether it was the
Indians who taught the colonists or vice versa (Sy, 1908;
Gross, 1920). At any rate, it was obviously the white man
who contributed the idea of replacing the axe-cut with a
hole bored into the tree and the many developments since,
which were made possible by the use of metal tools and uten-
sils. It is not always apparent, however, why some of these
practices evolved as they did. Such things as size and depth
of taphole and spout design seem to have become stabilized
quite early and quite uniformly throughout the maple region
(Hitchcock, 1928), although examination of the available
literature fails to show how or why.

There has been considerable interest in the physiology
and the economics of maple sap exploitation dating from.the
nineteenth century (Alvord, 1862), and many studies have been

initiated. One such study made by the Vermont Agricultural

Experiment Station (Jones, Edson, and Morse, 1903) was a
monumental piece of work and has servedas a source and refer-
ence for innumerable publications since. Unfortunately, most
of the publications on work of an experimental nature do not
give analyzable statistics. Some studies were made on a very
limited number of trees, and others show conflicting results.
In addition, considering the broad range of climatic varia-
tion met within the sugar maple region, there is some
probability that findings in one area may fail of application
in another.

For these reasons it was thought well worth while to
initiate a series of studies on maple sap production in
central Michigan. It was hoped to evaluate more closely
some of the factors concerned in practical "sugar bush"
operations. Such a study could also supply additional data
from.a different area within the sugar maple region to sub-
stantiate some of the findings in other areas.

It might be well to emphasize here that one self—
imposed limitation on this particular work was that it should
be restricted to a study of practices presently in use or
reasonable variations from these that might be readily
applied. It was thought that more quantitative information
“was needed about these accepted practices as well as the more
'basic research.

The factors studied included:

1.
2.
3.
LL.
5.
6.
7.
8.

Height of taphole from the ground

Spout design

Depth of taphole

Diameter of taphole

Plastic versus metal containers

Compass position of taphole on the tree

Several weather factors as affecting production
Number of tapholes per tree.

Each of these factors constituted a separate, though

sometimes superimposed, experiment. They are therefore taken

up separately in the following exposition. A statistical

evaluation is given for each. Examples of complete statis-

tical analyses for the sections are included in the appendix.

The analysis of variance was carried out for each experiment

with the usual "F" test, and, where applicable, the "t" test

was used for Specific comparisons.

CHAPTER II

LOCATION AND DESCRIPTION OF THE
EXPERIMENTAL AREA

The area chosen for this study is the northeast
corner of Baker Woodlot on the campus of Michigan State
University. Baker Woodlot is approximately sixty-eight
acres of northern hardwoods that were part of the original
grant to the institution in 1855. It is located in Section.1%
Township A North, Range 1 West, Ingham.County, Michigan (see
Figure 1, p. 6).

Fuelwood was removed for timber stand improvement
prior to 1928. In the winter of 1938-39, a harvest and im-
provement cutting removed approximately 1,000 board feet per
acre. A light selection cut was conducted in 19h7. In ad-
dition, some small portions of the woodlot were subjected to
experimental wildlife cuttings in 1950. The remaining trees
constitute a relatively dense, healthy stand of mixed hard-
woods. At present it is being managed for student instruction
and for research by members of the faculty and graduate stu-
dents in the field of conservation.

The experimental area with which this report is con-
cerned is about six acres in size and contains a very high
proportion of sugar maple (Acer saccharum, Marsh) with a

light mixture of basswood (Tilia americana, L.), elm (Ulmus

 

 

 

 

 

 

 

 

 

 

.z-e-q

 

 

 

 

 

LEGENLJ
0’0AK .‘PROJECl AREA
M-HARD MAPLE cc—CLEARCUT
BS’BASSW(.-OD P-POND
B‘BEECH S—SWAMP
C‘CHERRY I-SAPLING
MS-SOFT MAPLE Z-POLE
E‘ELM 3-MATURE

WP ”WALNUT PL ANTATION

Figure I

BAKER WOODLOT TYPE MAP

MICHIGAN STATE UNIVERSITY
EAST LANSING,MICHIGAN

Scahe

o 500 Iooofeet
L i LI

americana, L.), oak (Quercus spp., L.), and cherry (Prunus
serotina, Ehrh.). No other experiments were being conducted
in this portion of the woodlot. The maples range in size
from seedlings to about thirty inches in diameter at breast
height (four and one-half feet from the ground). The land
is level to gently undulating and consists of Miami and
Hillsdale soils. Maples of good tapping size are plentiful
though none had been tapped previously. The area therefore
represents a fairly typical sugar bush without the danger of
experimental results being affected by hidden injuries from
old tapholes.

CHAPTER III
PROCEDURE
I. APPLICATION OF TREATMENTS

The healthy maples of tapping size were selected and
numbered with paint. The diameter breast high (four and
one-half feet above ground), total height, crown length and
width, crown class, and apparent vigor were recorded. All
numbered trees from.ten to sixteen inches in diameter breast
high were classed as one-bucket trees, those from.sixteen to
twenty inches were classed as two-bucket trees, and all over
twenty inches as three-bucket trees. These limitations on
the number of buckets per tree conform to the standards widely
accepted by maple sirup producers.

The trees to receive specific treatments were chosen
at random with the limitations that (1) an equal number be
placed in treatment categories that were to be compared and
that (2) the major comparisons be made between trees having
an equal number of buckets per tree. This allowed a compari-
son of the production per tree without the further complica-
tion of comparing production per bucket. It also reduced the
spread in diameters of trees being used for the comparisons,
as the number of buckets per tree was determined by diameter

class. This was considered advantageous because of the well

9

recognized direct relationship between average tree size and
average production (Jones gt 31., Fox & Hubbard, Moore 93 31,,
McIntyre, and others).

All the trees were tapped twenty-four to forty-eight
hours in advance of the first favorable sap weather as fore-
cast by the U. S. Weather Bureau and the Michigan State
University Forestry Department. Such forecasting proved
very accurate on a short-range, local weather prediction, as
it had for several years (Robbins, 19h9). Recent develop-
ments indicate that the ability to tap maple trees immedi-
ately prior to the first tapping weather may have a
beneficial effect in addition to the obvious one of catching
the early flow and thus extending the season. In a study of
the effects of tapping date on yield, Douglass (1955) found
indications that tapping too far ahead of the actual sap
flow may reduce total production by unnecessarily exposing
the tapholes to infection by microorganisms.

The tapholes were made with a brace and regular tap-
ping bit, as is usually done in woods operations. Standard
Spouts or SpileS like those commonly used by Michigan pro-
ducers directed the sap from the tree to the bucket, with the
exception of a few oversize spouts which had to be made to
order. Plastic bags were tested as possible replacements for
sap buckets. These bags are available through a commercial

:maple sap equipment supply house and are being tried by some

' 10
producers. In all, 120 trees were tapped with a total of 168
buckets hung in each of four years.

The sap was collected between one and six p.m. each
day on which there was a significant run. Thecontents of
each bucket were weighed to the nearest tenth of a pound with
a standard dairy milk scale and the weight recorded. At the
same time a drop of fresh sap was examined in a hand refrac-
tometer to determine the sugar content. The refractometer
actually measures the per cent of total solids (Brix scale),
which includes minute quantities of proteins and minerals,
but corresponds very closely to total sugars (Bryan, 1910,
and Morrow, 1952). This estimate of the sugar content was
later used to convert pounds of sap to sugar yields for each
tree by multiplying the daily reading in per cent Brix times
the daily sap production in pounds. That trees vary consider-
ably in the "richness" of their sap is well documented (Marvhn
McIntyre, Morrow, Taylor, and others), so all of the data
were analyzed for differences both in total sap production
and total sugar production as affected by the treatments.

All data were transferred to I.B.M. cards. I.B.M.
machines performed the conversion of sap weight and per cent
Brix to pounds of sugar and provided the various summations
of data used in the computations.

One large maple tree was tapped on all four sides and

the spouts connected to four recording rain gages that were

11
adapted to collect sap but to exclude precipitation and trash
(see Figure 2, p. 12). These gages gave an indication of when
sap flow commenced and when it ceased on each side of the
tree. This information was supplemented by a recording
hygrothermograph in a nearby weather shelter (see Figure 3,

p. 12). These instruments gave graphic records of what was

happening at that location twenty-four hours a day.
II. EXPERIMENTAL DESIGN

It was decided to use the standard analysis~of—
variance technique with a two-way classification (Walker and
Lev, 1953) for each sap season. The data were then combined
for the four years of the study to increase the number of de-
grees of freedom and thus the sensitivity of the analysis.
The final analysis was, then, a three classification scheme
with each year corresponding to a replication.

Since the compass position of the taphole on the tree
was a major factor to be studied, using it in combination
with each of several other factors being studied proved con-
venient. This was done in the first four experiments which
were: (1) Height of Taphole and Compass Position, (2) Spout
Design and Compass Position, (3) Depth of Taphole and Compass
Position, and (h) Diameter of Taphole and Compass Position.
The other four "experiments" are reclassifications or re-

combinations of the data obtained from these same trees. The

 

Figure 2.
Recording rain gages adapted for sap
collecting.

 

Figure 3.
View of experimental area showing
weather shelter and rain gages.

12

13
fifth experiment is a comparison of yields from.the trees on
which plastic bags were used to catch the sap and those on
which metal buckets were used. The sixth is a summation of
the data from three experiments in which the treatments were
all balanced between the cardinal compass positions so that
other treatments could be ignored and the effect of compass
position brought out more strongly. The seventh is an at-
tempt to find some correlation between the yields of these
same trees and local weather factors. Number eight is a
comparison of yields from groups of trees of approximately
the same diameter but on which different numbers of buckets
were hung. Details of these eXperiments, as well as specific
literature reviews relating to each, are set out in the fol-
lowing sections of this report.

(For the purpose of this report, any reference to
statistical significance is intended to imply a probable
error of no more than five per cent. The use of the term
"highly significant" implies a probable error of one per cent

or less.

 

CHAPTER IV
HEIGHT OF TAPHOLE FROM GROUND
I. REVIEW OF LITERATURE

Jones, Edson, and Morse (1903), in a careful and
extensive study, found sap pressures to be greatest close to
the base of the trunk in maple trees. The quantity and the
quality (sugar content) proved to be poor both above and
below this point. The flow from roots that were tapped was
more variable and of a lesser total amount. Both the flow
and the sugar content decreased with height above the ground.
To quote the authors, "Changes that cause the flow to stop
seem.to start at the top of the tree and work down." They
therefore concluded that the normal tapping height was not
only the most convenient but probably close to the best por-
tion of the tree from.the standpoint of yield.

Bryan (1912) may have had reference to this work when
he recommended tapping at waist height or "the convenience
of the collector." McIntyre (1932) did some experimental
work in which he found tapping at breast height produced
both more and sweeter sap.

0n the basis of a study of two maple trees, Johnson
(1945) found "striking reduction" in sap production as spouts

were placed successively higher. However, he found this to

15
be offset somewhat by a tendency for higher sugar concentra-

tion.

II. EXPERIMENTAL PROCEDURE

If these findings--indicating that the most convenient
height was as good as, if not better than, any other--could
be substantiated for Michigan conditions, it would be a happy
circumstance. Maple sirup and sugar production at best calls
for considerable labor under adverse conditions. The hand-
ling of heavy buckets of sap either too close to the ground
or higher up on the tree would add discouragingly to this
burden. Therefore, an experiment was set up to compare
yields of sap and sugar from three different tapping heights.

Trees twenty inches or larger were chosen, as trees of
this size are considered capable of carrying three taps with-
out undue injury. For the first three years, three tapholes
were placed in a single quadrant of each tree so that all
‘were facing very nearly in the same direction. It was thought
thus to eliminate the variable of compass position from.com-
parisons of the yields from.different heights on individual
trees. The quadrant to be tapped, however, was chosen at
random so that an equal number of trees were tapped.in each
of the cardinal compass positions. Sap bags were used in-
stead of buckets.

The trees were tapped at one-, two-, and three-foot

16
heights. The tapholes were offset a few inches to avoid
having the collection bags interfere with each other and also
to minimize any effect one taphole might have in reducing the
flow from.cne nearby. Jones, EdsOn, and Morse (1903) found
that sap pressure in a taphole was lowered by opening another
hole as much as eight feet above or below the original tap
when the new hole was in the line of the grain of the wood
(not always vertical in a tree with spiral or twisted grain).
The distance over which this reduction was effective varied
directly with the amount of pressure in the taphole. Pres-
sure release was found to be over a much shorter distance
when the second tap was made out of the line of the grain of
the wood. 0n the other hand, these same men also found that
the flow of sap from the sides of the tapholes was almost as
high in volume as that from either the top or bottom of the
same tap, despite a considerably lower pressure. This latter
point seems to indicate that the effect of one tap on the
flow from another, though out of the line of the grain,
might be greater than would be accounted for by pressure
changes.

In order to add a factor of safety in this regard, the
experiment was revised the fourth year to the extent of plac-
ing the tapholes in different quadrants in each tree. The
same twelve trees were used. In 1956, however, the tapholes

were randomized over three compass positions so that there

17
were three quadrants with a different tapping height on each
for all twelve trees, giving a total of thirty-six tapholes
as in the previous seasons. This change in the method of
application produced no apparent change in the results for

the 1956 season as compared to the previous three seasons.
III. RESULTS AND DISCUSSION

The yield was consistently smaller from.the one-foot
height than it was from.the other two heights. This differ-
ence was statistically significant for three years of the
four that the experiment was carried on. The data for the
l95h season was of such a high order of variability that the
differences in mean yields did not show significance. How-
ever, the one-foot tap had a total yield almost as far below
the average for the other two heights, percentagewise, as it
did during the other three seasons. Figure A, p. 18, shows
graphically the yields in pounds of sugar for all four sea-
sons. The results for sugar and sap were practically
identical. A summary of the analyses of variance of sap
yields for each year is given in the Appendix, p. 98.

The deficit in yield from.the one-foot height is
thought to be directly attributable to the fact that these
taps were occasionally buried in snow and remained frozen
when the higher taps thawed out and flowed substantially.

This did not occur often or over long periods of time, since

 

.2 ogmam

 

 

 

 

 

18

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

one: _ 3mm _ the; Re.

N _ m N _ .m m . n w _ II 2
III a.
I ON
I on

I I L

I an
II 0*

.I. II I at
I on

w]

..II no

 

 

¢<wam no

NAG... n2... “.0 ...IOGI >m 2:...
‘ 1403.0. unis). “.0 94m;

19
the snows were not very deep nor persistent. Conditions
which caused sap to flow usually caused the snow to recede
rather rapidly, particularly around the trees where the
trampling during sap gathering was combined with re-
radiation of the sun's energy from.the dark bark of the
trees. However, a few noticeable lags were observed each
year with no indication that the flow was prolonged from
these spouts after it ceased in the higher ones.

It is recognized that the range of tree sizes as well
as the heights of tapping investigated were quite limited in
this study. The comparison of three different heights on the
same tree has some advantage in stratifying the samples and
thus reducing variability. This may or may not offset the _
disadvantage of restricting the investigation to the larger,
three-bucket trees. There can be little doubt that investi-
gation of effect of height should be made for the smaller tree
sizes before any conclusions or recommendations can be
properly applied to them.

An extension of this study to include at least the
(four-foot level would be of value. This would not put the
buckets at too inconvenient a height. In areas of deep and
persistent snows an even greater height might be adopted
since much of the collecting would be done from an elevated
position on packed snow. Such conditions do not prevail in

the Central Michigan area, however, and four feet seems to

20

represent a practical limit unless yields are found to be

considerably better from.the elevated tapholes.

IV. CONCLUSIONS

The two-foot and three-foot heights failed to show any
significant differences. It is felt, therefore, that no rea-
son has been demonstrated for varying from the usual tapping
height, except perhaps to take the precaution of placing the

tapholes somewhat above the usual depth of snow for the

region.

CHAPTER V
SPOUT DESIGN
I. INTRODUCTION

Little has been written about the design of the spout
inserted into the taphole to guide the sap to the bucket,
except for the general observation that the old wooden Spouts
were very poor in comparison to the newer metal ones. The
wooden spouts were subject to splitting and, more important
still, to contamination by microorganisms that can greatly
reduce the quality of the product (Bryan, 1910) and perhaps
the quantity as well (Douglass, 1955). The metal spouts are
galvanized or, more recently, aluminum. They may be cast or
a rolled sheet construction. All are designed to block the
taphole only at the outer edge, or shoulder, being tapered
toward the inner end to avoid blocking the entry of sap into
the taphole from any direction. A hole or slit is provided
in the tapered end to allow the sap to flow into and through
the spout to the bucket without seeping or trickling down
the bark of the tree.

Most spouts are designed with a hook or projection
from.which the collecting bucket can be hung. The cast spouts
having an unpolished surface seem to cling to the tree better

tha11 the smooth rolled or the polished cast spouts. ‘When the

 

22
weight of the bucket pulls the spout loose, it generally re-
sults in the loss of the contents of that bucket as well as
any subsequent flow before the accident is discovered. Aside
from.such losses, it is difficult to see what difference the
type of spout could make on sap production. It is still more
difficult to see how it could affect the sugar content of the
sap. However, producers in different areas do contend that
one particular spout will produce better than another, and
there have even been claims that the sap from.cne is richer
than from another. There is no general agreement as to which
is best, so it was thought that some investigation of this
problem.should be made. An experiment was therefore set up
to test for differences in total yield of maple trees as

affected by Spout design.
II. EXPERIMENTAL PROCEDURE

It was decided to test three standard spout designs
commonly found in Michigan sugar bushes andone new aluminum
spout just recently available (see Figure 5, p. 23). Random
combinations of these four designs were distributed over
twenty-four two-bucket trees, and care was taken that the
same spout design was not assigned twice to the same tree.
These spouts were also assigned to one of the four cardinal
compass positions by a randomizing procedure. The total,

then, was forty-eight taps with spouts of each design appear-

ing in each quadrant three times.

 

Figure 5.
Spout designs compared for yield.

23

 

ELL

Under usual sugar bush procedure there might have been
reason to suspect that the yields would be seriously affected
by the loss of sap when spouts pulled loose. This would.mean
the loss of considerable sap, as the buckets are visited only
when there is sufficient volume over the entire sugar bush to
justify a collection. This experiment was so designed that
such losses would be held to a minimum. Daily collections
were made whether the flow was heavy or light, and daily
visits were made to the experimental area to gather weather
data. The experiment was, therefore, usually under observa-
tion twice a day. Any buckets that were blown from.the hooks
and any spouts pulled from the trees were quickly replaced.
Consequently, the data on total yield from.each Spout design

are relatively free of complication by this factor.
III. RESULTS AND DISCUSSION

The yields in pounds of sap from.these four spout de-
signs are presented graphically for all four seasons, 1953
through 1956, in Figure 6, p. 25. The complete analysis may
be found in the Appendix, pp. 99-102. In this experiment no
significant differences were demonstrated between the yields
from.the spout designs either for the four seasons combined
or for any season individually. The results in terms of

sugar were almost identical with those in terms of sap.

 

25

.0 ohsmfih

 

 

 

 

 

NE 23. can} 1434.4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

zwfimo ...—Loam >m n_<m ...O 04m;

I 000

II ooh

IIOOO.

I Oom—

II 000—

l on».

I OOON

II can“

I coon

I Oahu

m<m no
nozaom

 

26
IV. CONCLUSIONS

Judging from these results, any choice of spout
design should be made on the basis of price, convenience,
retention in the taphole, or some other factor, rather than

yield of sugar or sap.

 

CHAPTER VI
DEPTH OF TAPHOLE
I. INTRODUCTION

How deep to make the tapholes for maple sap has been
of interest to maple sap products producers and experimenters
for over fifty years. Tapping deeper than necessary has two
rather serious consequences. With each additional inch of
tapping, it becomes more difficult to drive the bit into the
wood and also to clean the chips from the hole. Therefore,
time and labor expenditure are increased rapidly. These fac-
tors may not loom so large where power drills are used, but
any increase in cost of production is important, whether it
be gasoline for a back-pack motor and additional drill bits
or hand labor with brace and bit.

The second consequence of unnecessary tapping depth is
an increased possibility of injury to the tree. All tapholes
produce a certain amount of discoloration and decay, not only
around the hole but up and down the tree for several feet
(McIntyre, 1932; Robbins, 19h9; and others). The effect is
relatively limited laterally so that the resultant area of
damage after a few years resembles a flat, irregularly shaped
plate situated radially approximately the depth of the tap-
hole and just a little thicker laterally than the diameter of

the hole. Not only is the size of the damaged area enlarged

28
by increased depth of tapping, but also there is a greater
probability of introducing heart rot. With careful tapping
methods, maple trees heal over quite readily on the outside
of the hole and the damage to the sapwood seems to progress
slowly, even though with time the injury spreads in all di-
rections. The butt portion of heavily tapped maple trees is
almost always discarded when these trees are cut for lumber.
Unnecessary injury may mean not only the untimely loss‘of a
good sap producer but also the loss of additional income if
an additional section of the valuable butt log must be

discarded.

II. REVIEW OF LITERATURE

Jones, Edson, and Morse (1903), whose work has been
cited previously, tested numerous trees by means of Special
spouts designed to help solve the problem of proper depth of
tapping. These spouts were actually a series of tubes within
tubes so designed that the sap from each inch and a half of
depth was trapped and collected separately. They used taps
from three to six inches deep with these Spouts and found
that the amount of flow increased with depth but at a rapidly
decreasing rate. More than four~fifths of the total volume
of sap was obtained from the first three inches, and they
found that the first inch and a half yielded the best sap.

They concluded that storage of carbohydrates in the tree

29
takes place from.the outside toward the center and that
stored food is drawn on in this order. This would account
for the low quality (sugar content) of the sap from deeper
in the tap.

These men used a similar technique to insert manometems
for reading sap pressure in the outer two and a half inches
and the inner two and a half inches of tapholes and found
that the sap pressure was greater near the surface than far-
ther in the taphole. This was in line with the considerably
greater volume of sap obtained from.the outer portion of the
tree. Jones and Bradlee (1933) did some detailed experimen-
tation on the carbohydrate contents of maple trees, excising
portions of thetrunks for chemical analysis. The fact that
they found little sugar in the inner wood as compared to the
outer helped to substantiate the earlier work.

Fox and Hubbard (1905) reported that most of the sap
was produced from the first inch and that sap from.deeper
holes produced darker and considerably inferior sirup. Ex-
periments by Chittendon in Michigan (1919) indicated that the
first inch and a half produced most of the sap. In a later
'work, Cope (19h9) compared four-inch taps with two-inch. He
found that his four-inch taps produced 20 to 30 per cent more
sap. Then, in a test of two trees with a specially designed
spout, he found that approximately 30 per cent of the sap

came from the deeper two inches with "no appreciable

 

30
difference in sugar content."

Considering the fact that heartwood is not capable of
translocating liquids, Bryan (1912) probably hit on one of
the primary reasons for disagreement as to the best depth to
tap when he said it depended on the depth to heartwood. He
recommended an inch and a half to two inches maximum depth,
with the admonition to stop boring when dark shavings indicat-
ing heartwood appeared.

In old virgin trees such as were often tapped in
earlier years, it is doubtful if there is much more than an
inch thick band of sapwood under the bark. 0n the other
hand, young open-grown stands of second growth maple, such as
are being tapped now, may have tappable trees with several
inches of sapwood. The questions then become: how much of
this sapwood will yield good sap? and, is this increase in
yield sufficient to offset any additional injury to the tree?
The following experiment was set up to give some quantitative
information about this for the growth conditions found in

Baker Woodlot.

III. EXPERIMENTAL PROCEDURE

It was decided to test three tapping depths—~two, foun
and six inches. Forty-eight one-bucket trees were used, di-
‘Vided at random so that sixteen received each treatment. The

method.of tapping and the equipment used were standard for

31
commercial operations in the region except that the greater
depths of tapping demanded an extra-long wood bit. The holes
were seven-sixteenths of an inch in diameter. The six-inch
tapping proved very laborious and time consuming. The trees
were young and relatively vigorous,-and only a few of the
taps to this depth showed any signs of having reached the
heartwood. The side of the tree to be tapped was again cho-
sen at random.with an equal number of trees in each treatment

category being tapped in each of the four cardinal compass

quadrants.

IV. RESULTS AND DISCUSSION

The total yields of sap for each tapping depth during
the four seasons 1953 through 1956 are presented graphically
in Figure 7, p. 32. For the first three years the figures
indicated a definite superiority of the four-inch tap over
the others. In 1953 and 1955 the difference between four-
inch and two-inch taps was statistically significant, and it
was very nearly so in 195A. In 1955 the superiority of the
four-inch over the six-inch tap was statistically significant
as well. The author feels that these differences are totally
unreliable. Unfortunately, the assumption was made that the
original randomization and the relatively large number of
trees in each treatment group had resulted in samples of the

population with equal mean yield capacity. Therefore, the

 

 

 

 

 

 

1%... 32 Ward moo. _ _

 

 

 

 

32

I go:
I coo:

. ..w .-v u-N . rum =¢ 3N :0 2* SN «.0 aw 3N I OOO.~

I oo.o.~

 

 

 

 

 

 

IFI I I coo.»

 

 

 

 

I I coo.»

 

I coo...

 

 

 

I 60......

I 606....

 

 

Ll

I come

who: m4... Huh”
...—0 IHn—ma >m m<m “.0 mozaon— Z. 04w;

 

 

33
treatments were not re-randomized each year. Instead, the
same treatments were repeated on the same trees each year.
This was not rectified until the fourth year, even though no
reason could be ascribed for the fact that the six-inch taps
yielded consistently and significantly (one year at least)
less than the four-inch. Had these tapholes reached heart-
wood consistently, it would be expected that the Six-inch
tapholes would yield very little more than the four-inch.
That they should yield significantly less is more difficult
to explain.

Under any experimental design there is always a possi-
bility of sampling error. In a population as widely variable
as sap-producing maple trees the possibility of tying the
results to initial differences in the samples should not have
been overlooked. For the fourth season the same forty-eight
trees were used, but the treatments were re-randomized over
theme The results justified the hack of faith in the pre-
vious figures. The differences were far from being statis-
tically significant, although this time the Six-inch tap took
the lead as it is suspected it should if there are any real
differences and if more sapwood was exposed by the deeper tap.

The statistical analysis for the 1956 season is pre-
sented.in the Appendix, pp.103-105 as an example of the
statistical technique. In all cases the results for total

Yield of sugar were very nearly identical with those for

31L
yield of sap, so only the analysis for sugar is given.

Since the results for 1954 and 1955 are considered
suspect, the usable data on depth of tap are limited to two
randomizations. There is also legitimate reason to question
whether the first samples were truly biased and, if so, whe-
ther the last year's sample was any better than the first.
For this reason, considerable caution must be exercised in
arriving at any conclusions. The first three years could per-
haps be given equal weight, 12 3323, with the last year. Thus
would tend to cancel the apparent differences found in the
first randomization since the relative positions were re-
versed. Such manipulations were considered of little value.
The net result is still a lack of sufficient evidence on
which to soundly base any conclusion. Further study is

indicated.

V. CONCLUSIONS

The results for the 1956 season are more nearly in
line with those obtained in other experiments. The differ-
ences, while generally being in favor of the deepest tap,
were hardly sufficient to warrant additional damage to the
tree. Therefore, at least until further data are obtained
proving otherwise, two inches would appear to be sufficient
depth to secure a high proportion of the potential yield with

a minimum of injury to the tree. Any sugar left in the tree

 

 

 

35

by conservative tapping can hardly be said to be wasted,
since it will be available to the tree for growth the fol-
lowing growing season, thus increasing the potential of the
tree to store sugar for the next sap season.

One conclusion that may be drawn from.this work is
that a sample of this size is not sufficient for reliable
results without more careful stratification and/or replica-

tion.

 

 

 

CHAPTER VII
DIAMETER OF TAPHOLE
I. INTRODUCTION

Apparently little experimental work has been done on
the diameter of the taphole for collecting maple sap. This
may well be due to the early and general recognition of the
damage done by opening up the bark of the tree. Historians
of the maple industry have described the early crude methods
used in obtaining the sap, which often killed the trees or
damaged them severely (Bryan, 1912; Gross, 1920; Hitchcock,
1928; Huffman gfi,§1., 19h0). They describe the method of cut-
ting a box in the tree itself to catch the sap, the axe
serving to make both the wound that released the sap and the
container that collected it. The introduction of the iron
auger was a tremendous improvement, but it still opened up a
hole from one to two inches in diameter. Steel bits came
into use around 1800 (Hitchcock, 1928). The most common
sizes then, as now, seem to have been from.three-eighths to
one-half inch, though examination of the literature fails to
disclose why these sizes were so quickly and so widely
adopted.

The desirability of rapid healing over of the opening

is stressed in later work. Jones g3 31. (1903) tested several

 

37

:sizes from.three-sixteenths to three-fourths of an inch.
Gflaey found that the larger holes produced more sap and that
efill were equal in sugar content. They concluded, however,
tuiat the maximum.allowable time for the wound to heal over
unas two to three growing seasons. Others have said it should
hue accomplished in one or two growing seasons (Fox and Hub-
tuard, 1905; Bryan, 1912). A11 seem to be in agreement that
tkrree-eighths to one-half inch is a good range in which heal-
iiig may be completed in a reasonable time without sacrificing
too much sap yield.

A few recommend reaming the taphole one-eighth inch
lixrger in diameter if the flow is interrupted by a warm
spell followed by good sap weather (Bryan, 1912; Chittendon,
15919). This may reactivate dry holes and is much less injur-
ixyus than the practice occasionally used of drilling a new
taphole beside the old. This latter practice tends to ex-
haust the area of fresh, uninjured sapwood available for
future tapping. New tapholes which intersect the discolored
or injured wood of an old tap seldom flow well or long
(MCIntyre, 1932; Robbins, 1919; and others).

Conservative tapping procedures that allow the wound
POIMBal readily and limit the number of buckets per tree can
extend the useful life of the trees. As soon as a reasonably
thick layer of sapwood is laid over the wound, as it will be

by a vigorous tree that is not "bled" too heavily, this

 

 

38
portion of the tree again becomes available for tapping
(McIntyre, 1932; Robbins, 19h9). Progressing around the tree
conservatively could make the natural life span of the maple
the limiting factor in its sap-producing life. The sugar
bush in Sanford Woodlot on the Michigan State University came
pus has been in almost continual production for over thirty
years, thanks to such conservative tapping, with little evi-
dence of serious damage.

The following experiment was set up to obtain some
quantitative data about the effects of varying the size of

the taphole.

II. EXPERIMENTAL PROCEDURE

It was first decided to test taphole diameters of
sevenrsixteenths, eleven-sixteenths, and fifteen-sixteenths
of an inch. Twelve one-bucket trees were chosen at random
for each of the three treatments (or diameter sizes), giving
a total of thirty-six trees tapped. As in previous experi-
ments, the quadrant to be tapped was also chosen at random,
so that an equal number of tapholes of each size were located
in each cardinal compass quadrant. Seven-sixteenths is a
standard size tap and one of the commercial Spouts was satis-
factory. For the larger holes, special spouts had to be
designed and made to order. These were turned from a one—

inch aluminum rod, bored and fitted with a small copper tube

 

39
(see Figure 8, p. NO). The copper tube projected beyond the
aluminum in order to conduct the sap cut away from the blunt
shoulder of the spout where it had a tendency to seep back
under and run down the bark, instead of dripping into the
bucket. A small lump of solder under the tip of the copper
tube also helped the drops to form.and fall free. These
spouts were not entirely satisfactory. Being of smoothly
turned aluminum, they tended to slip out of the tapholes too
easily. In addition, the one-inch aluminum.rod was so nearly
the size of the fifteen-sixteenths inch holes that there was
the risk of driving the Spouts a little deeper than the camr
bial layer, exposing a bit of sapwood and allowing some of
the sap to seep out around the Spout and run down the trunk
of the tree. Little sap was lost in this manner, and it is
felt that the effect on the results is negligible. However,
spouts made from rods larger than one inch in diameter would
eliminate this worry.

During the growing season following the first experi-
mental Sap collecting, the rate of healing of the tapholes
was observed. It was apparent that the fifteen-sixteenths
inch holes would be far from covered ever, even in three
years. The injury was causing some splitting and sloughing
of bark around the edges of the holes. For this reason it
was decided to limit the testing of these largest holes to

two years and substitute a size at the other end of the range,

 

 

ho

.on pSomm pudendum one Am one <v cos ESCHESHo Soon ponds» mpzomm oNHmao>o

A

.m endwam

 

    

  

M

 

 

 

hl
smaller than the seven-sixteenths tap. Since three-eighths
was commonly mentioned in the literature, this size was cho-
sen for use in place of the largest. The fifteen-sixteenths
taps were tested during the 1953 and the 1955 seasons. The
six-sixteenths (three-eighths) taps were tested during the
195A and the 1956 seasons. Both were compared to the seven-
and the eleven-sixteenths inch taps, which were used all four
years.

Unfortunately, as in the previous experiment, the
treatments were repeated in the same trees each year for
three years, the seven- and eleven-sixteenths taps being
assigned to the same trees all three years. The fifteen-
sixteenths inch taps were reassigned to the third sample
group of trees after the 195u season during which those
trees had been assigned the six-sixteenths size. The re-
sults for the first three years, therefore, are confounded
with any initial differences that may have existed in the
samples and for this reason are suspect. The treatments
were re-randomized over the thirty-six trees for the fourth

season to rectify this.

III. RESULTS AND DISCUSSION

The total yields in pounds of sap are given by tap-
hole size for all four years in a graph on page AZ (Figure 9%
The apparent superiority of the eleven-sixteenths inch taps

 

 

 

N2

mom.

 

 

 

 

 

mumm_

 

mm.

 

 

......x. ..

 

 

. o enema

.vmm.

 

 

:w-\: .

mum:

 

 

 

 

am.

 

 

 

 

Gum; n_<m wax—<5— Q24.
MAO... m4... ...O .GMHMZSQ

III oov
.IIIAUOQ
III cow.
.III 000.
.lIIguoou
III OO¢N

III OOQN

m<m no
mozaoa

 

 

 

113
may well be an initial superiority of the sample trees
receiving the treatment. On the other hand, the differences
in 1953 and 1955 were not statistically significant at the
five per cent level and so may be dismissed as sampling error
in any event. The analysis of the data for these years is
presented in the Appendix (pp. 106-108).

The analysis for l95h indicated that the yield of the
six-sixteenths inch taps was lower than that of the other two
sizes by an amount that was significant at the one per cent
level. However, when the treatments were re—randomized and
repeated in 1956, the relative yields of the different sizes
were reversed. Although the differences in the mean yields
in 1956 were not statistically significant, this reversal
counterbalanced the effect of the l95h data so that the com-
bined analysis for the two years (see Appendix, pp. 109-111)
showed no significant differences in yields by taphole size.
Furthermore, this reversal indicated the danger in relying

on such a small sample.
IV. CONCLUSIONS

The first and foremost conclusion must be that no sig-
nificant differences have been demonstrated in yields from
the various sizes of tapholes tested in this experiment.

Secondly, it may be concluded that a larger sample or

more careful stratification would be required to demonstrate

HI III..I -,

1411,

any differences that might exist. However, the failure to
find significant differences is encouraging in itself. The
samples were relatively large in Comparison to many maple sap
experiments, and the repetition of the experiment over a
period of years has given useful indications of what can be
expected. Therefore, one may feel free to recommend the use
of the smaller sizes of bits or drills and thereby minimize

damage to the trees, unless more evidence is accumulated to

the contrary.

 

 

CHAPTER VIII

TYPE OF CONTAINER

I. INTRODUCTION

A recent development in the maple sap industry has

been the introduction of flexible plastic bags to catch the

sap, instead of the metal buckets now generally used.

These

plastic bags offer some definite advantages over the buckets,

and more are claimed for them. One advantage is the fact

that they are translucent, nearly transparent, and the
of the contents can be noted quite readily without the
sity of walking up to the tree and removing the cover,
be done with metal buckets. This reduces time and leg

appreciably (see Figure 10 this page and Figure 11, p.

 

Figure 10.
Plastic sap bag.

height
neces-
as must

work

116).

 

 

116

    

Figure 11.
Metal sap bucket with snap-on cover.

 

Figure 12.
Emptying sap bag into collecting pail.

II?

To empty the plastic bags into the gathering pail they
need not be lifted from the spouts. Instead, the bottom.may
be pivoted upward by lifting one lower corner, allowingthe
contents to flow into the pail (see Figure 12, p. E6). This
operation is similar to the usual procedure with buckets in
which the bucket is tilted on the Spout to dump into (the
gathering pail without removing it from the Spout. Such a
system.is much quicker and more convenient than removing
buckets from.hooks on the Spouts to dump them, as is some-
times done.

There is some doubt as to whether the "welded" plastic
seams can stand the severe treatment of steaming which is
recommended for the metal buckets before starting the new
season. However, since the bags can be washed easily (in a
washing machine if available} steaming may prove entirely
unnecessary.

The plastic is subject to an occasional puncture or
Split seam, but sap gatherers can conveniently carry spare
bags folded in their pockets, and the damaged bag can be
cold-patched quite easily. How great the losses from this
sort of accident may be has not been tested sufficiently.

Bags are more expensive at present than buckets, and
their use would often require the replacement of all spouts
by a type that is designed to accommodate the bag. Buckets

are a familiar item and their durability well tested. Some

h8
are still giving service after thirty years of use.

Ice formation is seldom a problem in buckets. Occa-
sionally a bucket is split at the seam by the expansion of
ice formation, but usually the sloping sides of the bucket
prevent this damage. The ice can be dumped out--into the
gathering pail or onto the ground, depending on the operator's
views as to the value of "sap ice"--without much trouble. In
the case of the bags,ice formation is a more serious problem.
Since the bags are narrow at the top, large chunks of ice
will not pass through. Soft ice may be crushed in the bag
and then dumped, though the bag may be split in the attempt
if the ice is harder than it appears. Hard ice must be left
to thaw. Subsequent runs of sap may then overflow unless
frequently emptied, since the bag is partially filled by the
ice.

Other important claims made for the plastic bag are
that their use will materially increase both the quantity and
the quality of the sap produced. The latter claim, improve-
:ment of quality, seems rather well substantiated. It is
'widely accepted that the action of bacteria and other micro-
organisms causes sap to "sour" and produce a dark, strong-
.flavored sirup which is considered inferior to the light-
colored, mild-flavored sirup from fresh "sweet" sap (Edson,
‘Jones, and Carpenter, 1912; and others). Sproston and Lane

C1953) found substantially lower bacteria counts in sap

1L9
collected in plastic containers. Naghski and Willits (1953)
also found lower bacteria counts in sap from plastic bags and
developed the proposition that it was the sterilizing effect
of sunlight admitted through the nearly tranSparent plastic
that made the difference.

Sheneman and Costilow (1956) have been making a study
of the effect of bacteria on maple sap and sap production.
They question the importance of sunlight in controlling growth
of microorganisms in plastic bags on the basis (a) that di-
rect rays of sun would be limited under sugar bush conditions
in Michigan and (b) that the rays would not be expected to
penetrate more than a shallow layer. They have suggested
further that the possibility of an inhibiting effect from.the
plastic itself should not be overlooked. In any event, the
salutary effect of the use of plastic bags on the quality of
sap remains undisputed.

Little investigation of the effect on quantity of sap
produced appears to have been made, however. It seems rea-
sonable to suppose that if the plastic bags reduced the
contamination of the sap, they would also reduce the likeli-
hood.of contamination of the taphole. The bag is hung on
the spout by means of a keyhole—shaped opening at the back
and near the top of the bag. The end of the Spout is en-
closed.in the bag by a flap of plastic that falls down over

the neck of the bag, covering it and the Spout. The Spout is

50
thus subject to the same protection and/or inhibitory effects
of the plastic as is the sap, whether it be exposure to light,
volatilization of chemicals from the plastic, or some other
factor. Recent studies have demonstrated that microbiologi-
cal contamination of the taphole is an important factor in
lindting the flow of sap (Douglass, 1955; Sheneman and
Costilow, 1956). In view of these findings, a study of the
effect of the plastic on production seems appropriate.

The only direct reference to a comparison of yields
from.metal buckets and plastic bags that was found was the
study made by Douglass at Michigan State University (1955).
His work was concerned primarily with the date of tapping and
‘with the effect of biologically sterile techniques of tapping
to avoid contamination of the taphole. A side study was made
of the difference in yield from the two types of containers.
This study failed to show any significant difference.

The following experiment was set up for the 1953 sap

season and continued through the 1956 season.

II. EXPERIMENTAL PROCEDURE

Testing the type of container was not considered a
major phase of the maple sap investigation. Therefore, bags
were used only where it was considered that any effect they
miShthave would not be confounded with the major factors.

(Hue meant hanging bags on all the trees used for a particular

51
comparison and buckets on all the trees used for another
comparison. The data were analyzed after the season ended,
and only if the major factors showed no significant differ-
ences were the yields used to compare bag versus bucket.
Consequently, the numbers of containers of each type were not
always equal. One-bucket trees were used in all cases.

For the first three years the average yield from six-
teen bags was compared to the average from twelve buckets.
During the 1956 season, the bags and buckets were equal in
number but were not equally distributed as to the compass po-
sition in which they were hung. This fact would have made
little or no difference in the previous years, since the com-
pass position differences did not approach significance.
However, the unusual weather for the 1956 season caused a
significant drop in yield from the trees tapped on the north
side. This introduced a considerable bias into the study of
type of container, since there were eight metal buckets on
the north Side of trees but only four plastic bags. Balanc-
ing the numbers of bags and buckets by compass position would
have reduced the number in each class by too great an extent.
It was therefore decided to use a technique in the analysis
of the data that was applicable to uneven class numbers. An
adjustment was made for the expected class numbers, and the
adjustment was tested for interaction. Since no interaction

was indicated, adjusted values were computed and analyzed.

 

 

 

52"
This adjustment and analysis is presented in the Appendix,
pp. 112-115 0

III. RESULTS AND DISCUSSION

No significant differences were found between sap or
sugar yields as influenced by the type of container in which
the sap was collected for any one of the four years. It is
interesting to note, however, that for all four years the
plastic bags produced a greater yield than the metal buckets
(see Figure 13, p. 53). This might be construed as an indi-
cation that a tighter design with better control of variation
could Show these differences to be significant.

No difference was found in the quality of sap or sirup
from.these two containers. This may have been due to the un-
usual conditions of the experiment. Bags and buckets were
emptied daily if a flow occurred, regardless of the amount of
sap. There was, therefore, little chance of serious contami-

nation or "souring" of the sap in either type of container.
IV. CONCLUSIONS

This experiment has demonstrated no reason to change

the type of container for anticipated gain in production.
However, indications are that it would be well worth while to
follow results from the use of plastic in the future. An imr

provement in sap quality alone should be sufficient to make a

 

 

I I .. \.

. . .

. u : a

o
..
E
u
o
. L

 

 

\ag
\9

N\g

\\\;§§°
\v

\\ mew \\ g
\\-

w
oooooooooo
oooooooo
OON N
a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MEAN YIELD OF MAPLE SAP
BY TYPE OF CONTAINER

 

 

 

 

_ 5h
change profitable, and an increase in volume is still a good

 

possibility worthy of more study.

 

 

i

 

 

CHAPTER IX
COMPASS POSITION OF TAPHOLE
I. INTRODUCTION

The best side of the tree to tap to obtain the most or
the sweetest sap is a problem that has received considerable

attention for more than fifty years. The fact that there has

 

been no real agreement in this time argues that it must not
be a simple matter of one side being superior, or none being
superior, under any and all conditions.

The maple sap industry is restricted to a northerly
portion of the northern hemisphere, and activity is re-
Stricted to a season when the sun is still relatively low on
the horizon. The variation in amount of direct sunlight that
lean.reach different parts of the circumferences of trees, than,
is fairly extreme. This cannot help affecting the amount of
tirm during which any one side of the tree is thawed suffi-
ciently for sap to flow. It must also affect the amount of
time in which a particular side of the tree is exposed to
conditions conducive to "drying up" the tapholes. Interac-
tions of these conditions with other factors, such as length
or sap season, amount of direct sunlight during the season,
and other highly variable climatic factors, could be expected

to produce variable effects on the sap flow from the different

56

compass positions on the tree. It is not surprising, there- r
fore, that different observations or experiments, restricted ‘
as they must be to Specific weather conditions, are not in
agreement.

Any proof of the superiority of one side of the tree
over the others would be self-defeating in the long run. Re-
peated tapping of a particular quadrant of the tree tends to

concentrate the taphole injuries. This makes it increasingly

 

difficult to drill a hole which does not approach or inter-
sect the damaged area from a previous tap, with consequent
loss of production from the new hole. From the standpoint of
the life of the tree and continuous production of sap, the
ideal Situation would be that of a continuous shift around
the tree so that the entire circumference is used with a

Hdnimum of concentration in any quadrant.
II. REVIEW OF LITERATURE

In the literature examined there is little evidence of
EHY' strong preference for one particular side of the tree,
thougm.the south side has been ranked a little ahead of the
Others by some writers. Jones gt El- (1903) found yields
slightly in favor of the south side but not sufficient to
warrant the additional damage entailed in repeating taps on
one side. In later reports, Fox and Hubbard (1905) and

Chittendon (1919) favored the "sunny side" of the tree, which v

 

 

57
need not always be the south side, but is mwst likely to be.

Chittendon also claimed that the greatest flow came from.the
side with the heaviest branches. This, however, is in con-
tradiction to the findings of other research workers (Jones
gt 31., 1903; Moore, 32 31., 1951). No particular side of
the tree was favored by Bryan (1912), and McIntyre (1932)
failed to find any correlation between sugar content and the
side of the tree tapped. Several workers found evidence that
sap flow started earlier on the south side but continued
longer on the north, with the average yields being very
nearly the same for both (Huffman gt g1., 1940; Tressler and
Zimmerman, 19h2; and others). Some who found this relation-
ship used it in recommending that the first tapping of the
season be done on the south side and any later tapping be
done on the north (Moore gt 31., 1951; Morrow, 1955).

Two of the more recent studies contain indications
that the real problem may not be a matter of which one side
is best, but rather, which one Side may be most likely to
produce minimal sap flows. The data from.Robbins (19h8)
shows a remarkable equality of average flow from the east,
the south, and the west taps and a considerable drop in
yield from the north. Douglass (1955) found a Similar situ-
ation, though less pronounced than that shown by Robbins.
Neither of these men found statistically significant differ-

ences in yields by compass position. However, the apparent

58
deviation is worthy of note since it is in line with what
might be expected if the north side stayed frozen too often

or too close to the end of the sap season.
III. EXPERIMENTAL PROCEDURE

The first four experiments reported herein were set up
to include a comparison of yields by compass position as well
as one other major factor. The quadrant to be tapped was
chosen at random within the limitation of a balanced distri-
bution of both of the factors being studied. In all, 168
tapholes contributed data each year on the effect of compass
position. Each year the trees were tapped ninety degrees
clockwise from the previous tap. This assured that no taps
were repeated in any quadrant on the one-bucket trees. The
two-bucket trees had taps 180 degrees from each other. The
ninety-degree shift the first year operated to expose un-
'touched quadrants. In the third and fourth years care was
taken to tap in the proper quadrant, but at least three or
four inches from the old wound.

Each experiment was analyzed separately for possible
differences due to position and to the associated major factor.
The assumption was made that any experiment in which no sig-
nificant differences were found in the associated factor
(height, diameter, depth of taphole, or spout design) would

provide unbiased data on compass position alone, ignoring the

A C's-a.- new“- -..—ea. .. a A A

59
other factors. This increased the number of degrees of free-

dom of the analysis, presumably making it more sensitive.
IV. RESULTS AND DISCUSSION

The yields by compass position for each year are pre-
sented in a graph on p. 60 (Figure 1A). For the first three
years only one of the four experiments gave any indication of

significance due to compass position. This was the experi-

 

ment which was also concerned with height of taphole from.the
ground (see Analysis I, Appendix p. 98).

Each year, compass position proved significant, but
each year the one position which differed significantly turnai
out to be ninety degrees clockwise from the position that was
significantly different the previous year. The obvious ex-
planation is that the particular group of trees chosen to be
tapped in that quadrant the first year was not a representa-
tive sample; that is, the trees were not of equal potential
yield with those to be tapped in other quadrants. Since the
experimental design called for these same trees to be used in
the same experiment but all tapped ninety degrees clockwise
from the previous tap, it is to be expected that any differ-
ence in potential yield would Show up again, this time in the
new quadrant. This seems to have been the case for 19Sh and

1955.

For the fourth season, all of the experiments were

 

60

 

 

 

_ m a
— one.

 

 

 

 

 

 

 

 

 

 

 

 

H F _
fl one. _

 

 

 

 

 

.1:

ofipwfi h

 

 

 

 

 

 

 

r

a P
vow. _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_. zoEmE $3200 5

mdm MIT—<2 ....O 0.5; Z<m2

 

 

 

on

bk

00.

mm.

on.

n:

OON

oNN

OnN

m<m ....O
moaned

61
re-randomized to break up any such sampling bias. The result
was as anticipated. This time the position effect was not
significant, and the mean yields were very similar in rank to
the other experiments. It is probably safe to assume, then,
that these results were not due to compass position as such.
This same conclusion could have been reached statistically by
combining four years' data in one analysis using the years as
replications. The constant shift of the position that was
out of line each year would undoubtedly have cancelled any
significance. However, it was thought that more could be
gained by re-randomizing the experiments.

The yields of all 168 taps were combined in 1953. giv—
ing a total of forty-two taps in each quadrant. The analysis
still failed to show any significant difference between com-
pass positions. In l95h it was decided that 156 of the taps
were sufficiently free of bias with regard to compass posi-
tion to be combined in one analysis. This analysis, too,
failed to demonstrate any significant differences. Again in
1955 a combined analysis was run, this time with thirty-one
trees from each quadrant, and similar results obtained.

The 1956 sap season appears to have deviated from the
usual. A considerable portion of the sap flow occurred when
the air temperature was near or below freezing. Many of the
spouts on the south, the east, and the west sides of the

trees were exposed to direct sunlight at least part of the

 

62
day, and that portion of the tree warmed up sufficiently for
the sap to flow. The north side of the trees, receiving no
direct sunlight to raise the temperature, remained frozen.
While these north side taps flowed heavily when conditions
were right, and closed the gap somewhat, they failed in all
four experiments to yield as much sap as the other sides. A
contributing factor may have been the abruptness with which

the season came to a close once the weather broke.

 

Of the individual analyses, only one indicated that
this difference was significant; but when a combined analysis
was run, the inferiority of the north taps was found to be
highly significant (see Appendix, pp. 116-117). The combined
analysis included three of the four experiments with thirty-
three taps in each quadrant for a total of 132 buckets. The
fourth experiment was not considered comparable to the other
three and therefore was not included in the combined analysis.
The differences in yields by compass position in this fourth
experiment were similar in rank and degree to the others.

The reason for not including them was the presence of other
complicating factors, including a serious lack of balance
between compass positions.

It thus appears from these data and evidence from
other experiments (Robbins, l9h8; Douglass, 1955) that if
maple trees are tapped equally on all sides, some losses due

to north taps may be experienced in exceptional years. We

63
have no data to indicate how often these exceptional condi-
tions may be expected to recur. Apparently few investigators
have had occasion to note such conditions. Those who have,
felt that the magnitude of the loss was not such as to war-
rant any additional injury to the trees that might be caused
by concentrating taps on the other sides of the trees. Ac-
cording to Robbins' figures, the apparent reduction of yield
from what might have been expected of the north taps was
approximately 21 per cent. This amounted to a reduction of
total yield from all taps of only 5 per cent. The average
production figures for this experiment Show the reduction to
have been approximately 25 per cent for the north taps alone

and 6 per cent for the total yield.
V. CONCLUSIONS

Any differences in yield due to compass position of
the taphole have not proven sufficient to cause concern or
to justify concentration of tapholes on any particular side

of the tree.

 

 

 

 

CHAPTER X
NUMBER OF TAPHOLES PER TREE
I. INTRODUCTION

The number of buckets that may be hung per tree
without causing undue harm has been rather well agreed upon.
It is generally recommended that trees below eight or ten
inches not be tapped at all, that only trees larger than fif-
teen or sixteen inches have two tapholes, that only trees
over twenty inches have three, and so on. The range of
variation in such recommendations is very limited. The
assumption here seems to be that more tapholes will necess-
arily result in more sap being obtained from the tree, the
only concern being that no tree be too heavily drained by
having too many tapholes in any one season. This assumption
has apparently not often been tested or even questioned.

Since it is generally recognized that the average vol-
ume of flow Shows a straight line relationship with the
diameters of the trees (Fox and Hubbard, 1905; McIntyre, 1932;
Moore gt g;., 1951; and others), it is to be expected that
the trees with more tapholes will produce more sap. As long
as the number of buckets hung per tree is determined by dia-
meter, this will be true and will lend apparent support to

the original assumption. It has occurred to some, however,

 

 

 

 

65
to question and to test whether an increase in the number of
tapholes in trees of a given diameter will substantially in-

crease the flow of sap.
II. REVIEW OF LITERATURE

'Chittendon (1919) stated that incomplete data indi-
cated that substantial increase could be expected when a
second taphole is added to a tree but that a third did not
seem to add so much. Herbert (1923) claimed to have found
that maples with two tapholes yielded more than twice as much
sap as those with but one. No data were given on the number
of trees included in the test or the diameters. Hitchcock
(1928) analyzed data obtained by interviewing producers and
seems to have assumed that adding another taphole was more or
less equivalent to tapping another tree of the same size, as
long as the health of the tree was not endangered. McIntyre
(1932) reported a definite increase in yield by increasing the
number of buckets per tree. His data show an increase almost
proportionate to the number of buckets up to four or five
buckets on larger trees.

On the other hand Huffman gt gt. (1940) stated that
the yield of sirup per tree was not increased very greatly by
hanging more buckets per tree. Their information was based
mainly on questionnaires filled out by producers. Tressler

and Zimmerman (19h2) compared the production per bucket from

 

 

 

 

66
large trees having two or more buckets with that of smaller
trees having only one. They found that the larger trees aver-
aged less per bucket, but they apparently did not test trees

of the same or similar diameters against each other.
III. EXPERIMENTAL PROCEDURE

In the experiments thus far discussed, trees ranging
from one-bucket to three-buckets in size were utilized. The
usual diameter limits were followed, with two-bucket trees
starting at seventeen inches and three-bucket trees starting
at twenty-one inches. Clustered around these limits were trees
very nearly the same in diameter but assigned different num-
bers of tapholes. This afforded some opportunity to study
the advantage gained, in terms of sap yield, by adding an-
other bucket to the trees of approximately the same size.

Nine trees were found in the one-bucket and nine in
the two-bucket classes which varied only O.h55 inches in
their diameters. The eight smallest three-bucket trees aver-
aged 1.2h inches larger than the eight largest two-bucket
trees. It was decided that better information could be de-
rived by using all eight of these trees in each class rather
than reducing the number and thus making the sizes more
nearly alike. The average diameters were computed by the
"average basal area" method, since it is considered more ac-

curate than using the arithmetic means of the diameters

 

 

 

 

6?
(Chapman and Meyer, 1949). This computation is presented on
p. 118 in the Appendix. The statistical analysis of the
yields of one-bucket and two-bucket trees is presented on
pages 119 and 120 in the Appendix, and the analysis of two-
and three-bucket trees on pages 121 and 122.

It should be pointed out here that all of the three-
bucket trees were used in the experiment on height of taphole.
Therefore, each had one bucket hung at the one-foot level.
These lowest buckets produced significantly less sap than did
those at two and three feet. All of the two-bucket trees
were tapped between two and three feet from the ground. It
must be assumed,then, that the additional bucket did not
contribute its full quota to the production of the three-
bucket trees. This introduces a bias in favor of the two-
bucket trees. There is no way of knowing from the data
obtained whether this is in any way offset by the fact that
the three-bucket trees are 1.2h inches larger on the average
--a bias in the opposite direction. Nevertheless, it was
considered a good opportunity to check roughly the assumption
that the increase should be nearly proportionate to the num-

ber of buckets.
IV. RESULTS AND DISCUSSION

The total yield from the one-bucket trees was only

1 per cent less than that from the two-bucket trees in the

 

 

 

 

68
same size group. The two-bucket trees produced 85 per cent
as much sap as the three-bucket trees in their size group.
Neither of these differences proved statistically significant.

It is unfortunate that so few trees were available for
these comparisons. Since only two isolated diameter classes
of the range of sizes tapped could be used-~and the compari-
sons in one of these were confounded by another significant
variable-~the results cannot carry much weight. However, the
comparison between one-bucket and two-bucket trees should be
valid since the yields of all the trees that were of similar
size were included in the calculations without any further
selection or known bias. In addition, the two-bucket and
three-bucket trees showed much less difference in yield than
would be expected if the addition of another bucket was even
nearly equivalent to tapping another tree.

In the light of the scarcity of data on the problem,
this evidence Should probably not be discarded. It should,
however, be supplemented by further, more carefully designed

investigations.
V. CONCLUSIONS

On the basis of very limited data, the results of this
experiment indicate that the controlling factor in deciding
the number of tapholes to use per tree might possibly be the

amount that the buckets would be required to held between

 

 

 

 

 

69
collections, rather than anticipated additional yield. The
slight additional yields obtained in this instance by the
increase in number of tapholes would not offset the addi-

tional damage to the tree.

 

 

 

CHAPTER XI
WEATHER AND SAP FLOW
I. INTRODUCTION

Producers and investigators alike find it easy to
agree on the conditions prerequisite to a flow of good qual-
ity sap. The temperature must remain at or below freezing
for a while and then rise sufficiently to thaw and allow the
sap to flow. Beyond this, the actual mechanism of the sap
flow and the conditions that determine what the quantity and
quality for a given season will be remain quite clouded. If
more were known about these conditions, it might be possible
to better anticipate needs for the sap season.

In maple sap production, as with other agricultural
crops, the ability to predict the probable yearly production
Would be a valuable asset. It would affect many decisions
about such things as fuel and labor requirements, operating
capital, equipment, and perhaps whether to operate the sugar
bush at all. Many opinions and suppositions can be found
concerning weather conditions that lead to a good sap season,
but little data are available on any phase except that of the

effect of alternate freezing and thawing.

 

 

71
II. LITERATURE REVIEW

Jones gt gt. (1903) found that the pressures asso-
ciated with sap flow in maple trees developed only when the
temperature fluctuated around zero degrees centigrade. A
pressure rise was always preceded by a rise in temperature,
although the pressure response was not always proportionate
to the difference in temperature. The "tension" in the tree
appeared to be largely the resultant of two variables, the
minimum temperature of the preceding night and the maximum
temperature during the day.

Jones and Bradlee (1933) found that the more often the
temperature fluctuated between freezing and thawing, the bet-
ter the sugar season. Their data also indicated that cold
weather during the dormant season increased the chemical ac-
tivity in the trees so that more starches and other carbohy-
drates were converted to sugars. Relatively warm dormant
periods would thus result in lower production, since more of
the stored foods would be retained as insoluble starches and
hemicelluloses.

Many workers since have theorized on the specific ef-
fect of freezing and thawing (Stevens and Eggert, 1945;
Marvin, 1946; and others). Robbins (1949) put this know-
ledge of the prerequisite conditions for sap flow to good use
when he developed an accurate system of forecasting sap wea-

ther. Some of the more recent work indicates that sap flows

 

-._._ -¢_ .

 

 

 

 

 

72
may be stimulated by increases in temperature starting from
as much as seven degrees centigrade above freezing (Marvin
and Greene, 1951; Taylor, 1953; and others). Marvin and
Erickson (1956) reported better correlations with sap flow
when using twig temperature than those found with inner bark
or xylem temperatures.

Since the sugar content of maple sap is a storage
product, the result of photosynthetic activity above that re-
quired for the growth of the tree during the current growing
season, the assumption is often made that the conditions of
the growing season directly preceding the sap season in ques-
tion will determine the amount of sugar available. The
difficulty lies in trying to identify specific conditions
that might be correlated with sap flow.

Jones gt gt. (1903) stated that seasonal variation in
leaf area was related to sugar content of the tree. They
also had an opportunity to observe a sugar bush that was de-
foliated by caterpillars and found that during the following
season the sap had a low sugar content. Sy (1908), in his
historical account, made reference to the amount of starch
stored depending on leaf development and sunlight during the
preceding growing season but offered no data to substantiate
it. McIntyre (1932) found sugar production to be directly
proportional to leaf area and amount of sunshine received.

His data, however, dealt with comparisons between trees of

._,_‘—.. 1‘-”

.1.x_:._‘ “-3: .

_ “<e—£—‘a .a 4-.. - .

 

73
different degrees of crown development rather than with
yearly variations as affecting individual trees.

Herbert (1924) attempted to correlate sap production
with mean monthly weather records. Curves of annual sap
yields for ten years were superimposed on curves of tempera-
ture, precipitation, and amount of sunshine during specific
seasons of the year as obtained from the monthly weather re-
ports. The curves of production and rainfall between May and
October followed each other rather closely with one serious
and unexplained deviation. The curves of production and per
cent of possible sunshine during sap season showed a gener-
ally similar trend, but again, important exceptions were
noted. The closest agreement in all of these curves seemed
to have been between production and precipitation just pre-
ceding and into the sap season, January through March. This
would appear to bear out the conviction of many producers
that plenty of snow means a good sap season. Huffman gt gt.
(1940) reported that a heavy snowfall was conducive to good
yields and to high quality, but they related this to the
coldness of the weather as indicated by the snow rather than

to the snow itself.
III. EXPERIMENTAL PROCEDURE

It seems obvious that much more investigation must be

undertaken before the relationships between these climatic

ass—....

 

71L

factors and maple sap production can be clearly established.
This experiment was not designed with the study of weather as
a major factor, but in conjunction with the production experi-
ments reported previously some weather records were obtained
in the sugar bush. Other data were available from local wea-
ther stations. Considering the complexity and variability of
climatic factors, four years is very probably less than one—

tenth of the time needed to establish reliable trends and

 

relationships. However, an opportunity presented itself here
to obtain daily sap records and some daily weather records
from the same immediate area.

A weather shelter was set up in the eXperimental area
to house a recording hygrothermograph. A twenty-four hour
clock device was used with this instrument so that an accu-
rate reading could be obtained for any hour of the day. The
relative humidity figures were checked against those of the
nearest weather station which, fortunately, was less than a
mile away. Daily readings of barometric pressure and per
cent of total possible sunshine were obtained from the
Weather Bureau. Line graphs were constructed for these four
factors in conjunction with a histogram of daily maple sap
production each year in an attempt to bring out trends or
correlations. Two examples of these combined graphs may be
seen on pages 75 and 76 (Figures 15 and 16). Charts were

also prepared annually showing daily maximum and minimum

 

 

 

   

 

 

 

 

.l. :- 11.I.I_I 11""

 

   

 

 

 

77
temperatures in conjunction with daily sap production. These
charts are on pages 78 through 81 (Figures 17, 18, 19, and
20). For the first three years a record of mean daily wind
direction and velocity was included.

Weather Bureau data were used to construct a series of
bar graphs comparing total sap production with: (1) winter
snowfall (Figure 21, p. 82), (2) rainfall during the previous
growing season (Figure 22, p. 83), (3) total precipitation
for the preceding year (Figure 23, p. 84), and (4) length of

previous growing season (Figure 24, p. 85).
IV. RESULTS AND DISCUSSION

A study of the line graphs of relative humidity, per
cent of sunshine, and barometric pressure revealed no obvious
relationships to sap yields. Air temperature showed a very
Close correlation with sap flow until the last week of the
sap season, as was expected. Some interaction between air
temperature and sunshine was noted. On a few days when the
air temperature remained below freezing, some significant sap
flow was recorded as a result of the warming of the trees by
direct sunlight. Little can be said to have been added to
OUP understanding of sap flow by these charts.

From the graphs of maximum and minimum temperatures
it Was thought that perhaps a relationship could be estab-

lished between duration or severity of the freeze--as

 

 

 

 

 

 

 

TOTAL WINTER SNOWFALL _ _

COMPARED TO SAP PRODUCTION

INCHES
OF SNOW

 

 

N «stung

BSSSS§:\\\\\\\\\

uouenoosa 996:

_ \\\ \\\\\ _
\ ,

M ‘numous ...—....
\\\XX\\\\,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘UNHMMMHEHN
. ‘\

\\\\\\\\‘
'n snows SS-ZSGI
\\\\\\\\\,

 

 

go. o o o o o o o
z<<> o c> o o o o
30’ o o o o o q o
On. 0 no" 0 o" o no 0
0.0 c '0 IO N N - ‘-

Figure 21 .

 

 

, fl

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I” 3 1’ a .. 83
<2: I I l
g; gymmhnw
Z
‘II'I E \\\\\\\
(.9 O
E D \ \\\ \ \ \ \ \\x
° \\
g 8 \T{‘€"\°£¥\°{61\
F, “- \\\\\\\\..:.h>.:.er.:.>\
a. s'
(D < o
D a,
O U) \\\x\\\\\\\ E:
S o Newscast?
g I \\\\s
a’ a \ \ \\ \ \
- DJ
2
I E mean?“
j % \\‘ \ \ \ \\ \ \
E J
\
é ‘8 I I I» l ' i l

 

 

TOTAL PRECIPITATION FOR PREVIOUS YEAR

COMPARED TO SAP PRODUCTION

INCHES

Pouuos
0F SAP
40,000—

40

o I) o
n «I N 2 2

X\\\\\\\\h\
:::\‘ NOIicnaosa cseI P\
'KVY\\\\\Y\

 

 

 

 

 

 

 

 

 

\ \\\\\\\VY
\ NOIiuIeIoaud 99m }\
I\ \\\\\\\\Y\

 

 

 

 

\c:;\\‘\\\\\\x\s‘\
“\\\ NOIicnOOSd 996I
X \\\\\\\\\\

 

 

 

 

 

 

 

 

\\\\\ \\\\\ \
\ NOILVIIdIOBUd 996|
\ \\\\\\\ \\

 

 

 

 

\\\\\\\\\\
S:ESS;S;E§E;E:j NOIionaosa veal L\
\\\\\\\ \\\

 

 

 

\ \\ \ \Q\ \\_\
NOI1V1|d|O38d ESGI

 

 

 

 

\\\\\\X\\\
\[ NOIicnooud 296f1\\

\\\\\\\\\\\

 

 

 

\\\\\\\\\\

 

NOIIVIIdIOBUd ZSGII

 

 

3aooo-—
3dOOO——
zaOOO——
2dOOO——
Iaooo——
Imooo——

Figure 23 .

 

 

z
o
(D
4.
DJ
CD
0

- E
3
o
I):
<9
(0
D
9
>
DJ
0:
0.
LI.
0
I
'—
(D
Z
I.I.l
..I

 

COMPARED TO SAP PRODUCTION

POUNDS
OF SAP

)-

V

N O Q

 

 

 

 

 

S\\\\\\\\ was?“ ~

 

 

 

 

 

\§§§§VSRHESR}QQR\

 

 

 

x Warsaw? \\

 

 

 

 

 

 

‘\\\\\\\\\\
F\;:<:i3:;:s::\\ NOIionaoad 996! \\

 

Figure 24.

 

 

 

\\\\\\\\x
h\:<:i;3§3i;i;:\\\ uosvas seal
\\\\\\\Y\

 

 

 

 

 

\\ NOLionaosd ass:\\
\\\\\\\\\\\

 

 

 

 

\\\\\\\\\ a: 9:13:32

 

 

 

40,000—

3aOOO——-

O O O O O

O O O O O

O O O. 0 C1

o In 9 n O I
n N N - '-

86
indicated by the shaded area below the thirty-two degree
line-~and the sap flow following this freeze. No such cor-
relation was apparent. The closest relationship was between
maximum daily temperature and sap flow, the sap flows occur-
ring very nearly in proportion to the height to which the
temperature rose during the day. This is in line with the
findings of Jones EE.&I° (1903). All temperature relation-
ships were cancelled by the drying of the tapholes after a
prolonged warm spell. Even a return of freezing nights and
bright days failed to stimulate a renewed flow (see Figure 1%
p- 78).

No trends could be discerned for sap yield as com—
pared to precipitation either during the growing season or
during the entire year preceding the sap season, nor with the
length of the growing season. Perhaps if data were available
over a long period, some relationship might be revealed. Four
years can be only a beginning.

During the first three years of the experiment it
began to look as if total winter snowfall were very closely
related to sap production (see Figure 21, p. 82). During
the fourth year this seeming correlation collapsed. Further
study is needed to disclose whether 1956 was an exceptional
year in which overriding factors obscured the influence of
the snow cover or, indeed, whether the influence of snow

cover is of any real importance.

87

The possibility should not be overlooked that snow may
exert a beneficial effect in providing plenty of moisture at
the time of sap flow. Large quantities of water are removed
from the trees and presumably replaced from soil moisture.
During this experimental period an average of twenty-seven
gallons per tree was removed each season.

Another possibility worthy of investigation is that a
blanket of snow may prevent the soil and roots from freezing,
thus enabling the roots to more readily replace the water

lost when sap is withdrawn.
V. CONCLUSIONS

Little direct benefit was gained by the study of
weather factors. However, four years of data on daily wea-
ther and sap flow may serve as a basis for continued study of
these factors. It is hoped that such will be the case since
so little is known about the effect of weather and so many

Years of data will be required to properly fill this gap.

 

 

2.

CHAPTER XII
CONCLUSIONS AND RECOMMENDATIONS
I. CONCLUSIONS

Tapholes between two and three feet from the ground
produced comparable yields of sap. A reduction in yield
from the one-foot tapholes appears to have resulted when
these lower taps were covered with snow and remained
frozen considerably longer than the others.

Differences in the four conventional spout designs had
little effect on sap or sugar yields so long as the

spouts were not subject to being pulled loose by the
weight of the bucket and sap.

An increase in tapping depth from two to four and six
inches had little effect on production. Any gain would
appear to be more than offset by the additional damage to
the tree.

No statistically significant differences were established
in the yields obtained from tapholes ranging in diameter
from three-eighths to fifteen—sixteenths inches.

No Significant differences in yields due to type of con-
tainer, metal or plastic, were demonstrated. Consistently
larger yields from the plastic bags indicated a need for
further study despite the lack of statistical significance

in.this limited comparison.

 

 

6.

7.

l.

2.

3.

89
Tapholes on the north side of the trees may in some years
produce significantly less sap than those in other quad-
rants. This occurred once in the four years of this
study. The difference was not sufficient to justify
avoiding the north side at the risk of concentrating
taphole damage in the other quadrants.
Very limited data indicated that there may be little
advantage to increasing the number of tapholes per tree
beyond the number of buckets necessary to prevent over-
flow between collections.
A close relationship was noted between maximum daily
temperature and sap flow until prolonged warm.weather

caused the flow to taper off and cease.
II. RECOMMENDATIONS

Larger samples or more replication are indicated for a
study such as this, particularly if only one or two
years' data are obtained.

More careful stratification of the samples, based on
crown development (or other indications of vigor) or per-
haps on previous history of production, should increase
the sensitivity of the experiment for any given sample
size.

It would be of interest to obtain further data on height

of taphole, starting from a level somewhat above normal

90
snow accumulation and extending well over three feet up
the tree.

A study designed to test the retention of various spout
designs in the taphole would be of value. This character-
istic appears to be capable of more effect on yield of
sap than the design itself.

A more careful study of taphole diameter is needed. Such
a study might well start at nine-sixteenths of an inch
and take in several sizes below this, possibly to five-
sixteenths of an inch, to find the smallest diameter
commensurate with reasonably good yields.

Developments in the use of plastic bags should be fol-
lowed, not only from the standpoint of quality of sap but
also from that of yield.

Further investigation of the effect on yield of adding
tapholes to trees of a given diameter is indicated. This
study covered only two sizes and a limited number of com-
parisons in each. A complicating factor in one of the
sizes tested weakened the results still further.

A study of the effect of snow cover on sap and sugar
yield might well be extended to cover the possibility of
its contribution to soil moisture at the critical period
of sap withdrawal.

An investigation of soil temperature, with and without

snow cover, and its influence on sap yield is another

 

 

(III I‘IIIIII INVI\
n’. d ... III I
II .u: I .I I

,.

10.

91

possibility in this problem of snow cover.

ExaminatiOn of the literature and Observation of the un-
predictable variation between trees of apparently the
same size and vigor on the same or similar sites points
strongly toward more intensive investigation of genetic
factors. Some of the earliest writers noted the large
and unexplained individual variation between trees
(Alvord, 1862; Fox and Hubbard, 1905). Since then, many
have ascribed this variation to "inherent characteris-
tics" or a "genetic factor" (McIntyre, 1932; Moore, gt
gt., 1951; and others), and some investigations have
shown such genetic causation to be highly probable
(Stevens and Dunn, 1943; Morrow, 1952 and 1955; Marvin,

1933; Taylor. 1933).

LI TERATURE C I TED

 

 

x,-

LITERATURE CITED

Alvord, C. T. The manufacture of maple sugar. Serial set
of U. S. Documents. Serial 1168 (1862 & 1863), pp. 394-405.

Anderson, W. R., J. S. Ball, H. R. Moore, and Richard Baker.
Factors influencing yield and cost of maple syrup. Ohio
Farm.and Home Research. 34 (1949).

Bryan, A. H. Maple-sap sirup: its manufacture, composition
and effect of environment thereon. U. S. Department of
Agriculture Bureau of Chemistry, Washington, Bulletin 134,
1910.

. The production of maple sirup and sugar. U. S.
- Department of Agriculture, Washington, Farmers Bulletin 516,
1912.

Chapman, H. H. and W. H. Meyer. Forest Mensuration. New
York: iMcGraw-Hill Book Company, 1949, 522 pp.

 

Chittendon, A. K. Notes on maple syrup making. Michigan
Agricultural Experiment Station, East Lansing, Quarterly
Bulletin 1 (number 3), 1919.

Cope, J. A. Depth of tapping in relation to yield of maple
sap. Journal gt Forestry. 47 (June 1949), pp. 478-480.

Douglass, J. E. The effect of date of tapping on the yield
of maple sap from sterile and non-sterile tapholes. Unpub-
lished M. S. thesis, Michigan State University, 1955.

Fox, W. F. and W. F. Hubbard. The maple sugar industry.
U. S. Department of Agriculture Bureau of Forestry, Washing-
ton, Bulletin 59, 1905.

Grose, L. R. Maple sugar in colonial times. American
Forestry. 26 (November 1920). PP. 689-690.

Herbert, P. A. Maple syrup making: two buckets per tree
most profitable. Michigan Agricultural Experiment Station,
East Lansing, Quarterly Bulletin 5 (number 3). 1923,

pp. 139-1A1-

. The weather and maple sugar production. Michigan
Agricultural Experiment Station, East Lansing, Quarterly
Bulletin 7 (number 2), 1924, pp. 60-62.

9h

Hitchcock, J. A. The economics of the farm.manufacture of
maple syrup and sugar. Vermont Agricultural Experiment
Station, Burlington, Bulletin 285, 1928.

Huffman, R. E., S. H. DeVault and J. W. Coddington. An
economic study of the maple products industry in Garrett
County, Maryland. ‘University of Maryland Agricultural
Experiment Station, College Park, Bulletin 431, 1940.

Johnson, L. P. V. Physiological studies on sapflow in the
sugar maple, Acer saccharum, Marsh. Canadian Journal gt
Research. C 23'TDecember 1945), pp. 192-197.

 

Jones, C. E., A. W. Edson and W. J. Morse. The maple sap
flow. Vermont Agricultural Experiment Station, Burlington,
Bulletin 103, 1903.

.and J. L. Bradlee. The carbohydrate contents of the
maple tree. University of Vermont and State Agricultural
College, Burlington, Bulletin 358, 1933.

Marvin, J. W. Some factors influencing sap flow in Acer
saccharum, Marsh. Amer. Journal gt Botany. 33 (October

1914.67, p o 82“..

. A report of continuing botanical researches on
maple sap at the University of Vermont. Proceedings gt tgg
Second Conference gt Mgple Products. Eastern Utilization
Research Branch, Agricultural Research Service, United
States Department of Agriculture, Philadelphia, 1953.

 

 

 

. and M. T. Greene. Temperature induced sap flow in
excised stems of Acer. Plant Physiology. 26 (July 1951),
pp- 565-580.

and R. O. Erickson. A statistical evaluation of
some of the factors reSponsible for the flow of sap from
the sugar maple. Plant Physiology. 31 (Jan. 1956), pp.
7-610

McIntyre, A. C. The maple products industry of Pennsylvania.
Pennsylvania State College School of Agriculture and Ex-
periment Station, State College, Bulletin 280, 1932.

Mbore, H. R., W. R. Anderson and R. H. Baker. Ohio maple
syrup: some factors influencing production. Ohio Agricul-
tugal Experiment Station,Wooster, Research Bulletin 718,

19 1.

 

 

 

95

Morrow, R. R. Consistency in sweetness and flow of maple sap.
Journal gt Forestry. 5O (Fefiruary 1952), pp. 130-131.

. Early tapping for more quality sirup. Journal gt
Forestry. 53 (January 1955), pp. 24-25.

. Influence of tree crowns on maple sap production.
Cornell University Agricultural Experiment Station, Ithaca,
Bulletin 916, 1955.

Naghski, J. and C. O. Willits. The sterilizing effect of
sunlight on maple sap collected in a transparent plastic
bag. Food Technology. 7 (February 1953), pp. 81-83.

Robbins, P. W. Position of tapping and other factors affect-
ing the flow of maple sap. Unpublished M. S. thesis,
Michigan State College, East Lansing, 1948.

. Production of maple sirup in Michigan. Michigan
State College Agricultural Experiment Station, East Lansing,
Circular Bulletin 213, 1949.

Sheneman, J. M. and R. N. Costilow. Oral and written com-
munication, 1956.

Sproston, T., Jr. and S. A. Lane. Maple sap contamination
and maple sap buckets. Vermont Agricultural Experiment
Station, Burlington, Station Pamphlet 28, 1953.

Stevens, C. L. and Dunn, Stuart. Increasing the sugar con-
tent of sugar maples. Journal gt Forestry. 41 (December
1914-3), pa 922.

and R. L. Eggert. Observations on the causes of the
flow of sap in red maple. Plant Physiology. 20 (October
19A5). pp. 636—648-

Sy, A. P. History, manufacture and analysis of maple prod-
ucts. Journal gt the Franklin Institute. 166 (1908),
pp. 249-280.

Taylor, F. H. Operation of an experimental sugar bush. tag-
ceedings gt tgg Second Conference gg Maple Products. Eastern
Utilization Research Branch, Agricultural Research Service,
United States Department of Agriculture, Philadelphia, 1953.

 

Tressler, C. J. and W. I. Zimmerman. Three years' operation
of an experimental sugar bush. New York Agricultural Experi-
ment Station, Geneva, Bulletin 699, 1942.

 

 

96

United States Department of Agriculture, Agricultural Statis-
tics. 1955, p. 81.

Walker, H. M. and J. Lev. Statistical Inference. New York:
Henry Holt and Company, 1953, pp. 348-386.

 

 

 

 

 

 

 

1953
Total
Pos ition
Height
iji
Error

15229;
Total
Pos ition
He ight
PxH

Error

19

otal
Pos ition
Height
PxH
Error

19 6
Total

Po 5 ition
He ight
PxH
Error

ANALYSIS I

HEIGHT 0F TAPHOLE AND COMPASS POSITION

Degrees
Freedom

Sum of

Sguares

79.055
28,277
13,609

7,325
29.8h4

364.763
86,01
35.97
17,142

225,629

176.024
47.298
47.736
11,6h4
69.3u6

137.206
17.513
23,600
18,245
77,8u8

Me an

Sguare

9J£6
6.305
1,221

28,671
17,989
2.857
9,401

15,766
23,868
1,941
2,889

8:756
11, 800
4,561
2,883

"F I!
Value

7.35%
5.07%
.98

3.05*
1.91
.30

98

Value of
Fat 5%

3.01
3.40
2.51

3.01
3.40
2.51

3.01

 

 

 

99
ANALYSIS II
SPOUT DESIGN AND COMPASS POSITION
4 YEARS COMBINED)
A2311§1§_2£_Y§£l§££2
' A Degrees Sum of Mean "F" "F"
Sew—Ce wmmmw
Total 191 1,096,309
Year (Y) 3 115,601 38,534 7.45** 2.68
Spile (S) 3 17,515 5,838 1.13 2.68
YxS 9 23,412 2,601 1.95
Position (P) 3 20,304 6,768 1.31 2.68
ny 9 43.376 4,820
SxP 9 67,667 7,519 1.45 1.95
SxYXP 27 146,075 5,410 1.05 1.57
Error 128 662,359 5,175

(1953) (1954) (1955) (1956)
22x2 = 790,321 + 2,120,288 + 1,700,569 + 1,314,628 = 5,925,806

(1953) (1954) (1955’ (1956)
EX = 5.863 + 8,966 + 8,355 + 7,267 = 3o,LI51

C.T. = gsz = (30,45122 = 927,263,40 = 4,829,496.88
N 192 192

Total $5 =Ex2 - C.T. = 1,096,309.12
Year ss = (5.863)2 + (8,966)2 $68,355)2 + (7,267)2 - C.T.
= 115,601.16
SPout SS = (8,123)2 + (6.890)2 i8(7,852)2 + (7.585)2 - C.T.
= 17,198.07
Position 33 = (7,050)2 + (7,960)2 £8(8,228)2 + (7,213)2 - C.T.
= 20,250.23

 

 

 

 

...____.——'——— ‘

100
ANALYSIS 11 (continued)

Two-Way Table for Interactiog (SxY)

 

 

 

 

 

Year A B C D Total
1953 1:531 19368 1:466 1:498 5,863
1954 2.528 1.771 2.543 2,124 8,966
1955 2,260 1,980 1,969 2,146 8,355
1956 1,80g 1,771 1,814 .1,817 1,266
Total 8,123 6,890 7,852 7,585 30,A50
' D.F.,

Total SS for SKY = 59,828,49g - C.T. = 156,210.95 15

12

Year SS (as before) = 115,601.16 3

Spout SS (as before) = 17,198.07 3

Spout x Year SS (by subtraction) = 23,411.72 9

Two-Way Table for Interaction (PxY)

 

Year North East South West Total
1953 1:401 1,537 1,639 1,286 5,863
1958 2,379 2.530 2,012 2,045 8.966
1955 1.934 2:005 2.377 2,039 8,355
1956 1,332 1,881 2,200 1,842 1,266
Total 7,051 7,959 8,228 7,212 30,450
D.F.
Total 83 for ny = 60,105,33 - C.T. = 179,597.81 15
12
Year SS (as before) = 115,918-35
Position SS (as before) = 20:303-85 3
YxP SS (by subtraction) = 43,375.61 9

ANALYSIS II (continued)

Two-Way Table for Interactiqg (SxP)

101

 

 

Spout

De_siss $122172 Test So_u__w Te: Lesa
A 1,805 2,177 2,204 1,937 8,123
B 1,916 1,321 2,001 1,652 6,890
C 1,627 2,125 2,378 1,722 7,852
D 1.1% 2.232 14616 1.201 1.282
Total 7,051 7,959 8,228 7,212 30,450

Total SS for SxP = 5922131994 - C.T. = 105,486.48 .QIUJ

Spout SS (as before) = 17,515.26 3

Position SS (as before) = 20,303.85 3

SxP SS (by subtraction) = 67,667.37 9

 

 

102

ANALYSIS II (continued)

Three-Way Table for Interaction (SxPxY)

 

 

 

 

 

 

Spout
Year Spout North East South West Design Yearly
Design Totals Totals
A 439 411 362 319 1,531
1953 B 378 274 428 288 1,368 5 863
c 274 403 389 400 1,466 a
D 310 449 460 279 1.498
a 232 212 22% as; 8%:
195“ c 671 699 785 388 2:543 8’966
D 512 723 324 565 2,214
A £22 550 22% 443 5,530
B 312 77 9 0
1955 c 388 529 571 481 1,969 8’35;
D 475 614 419 638 2,146
A 32? i6; 106 422 1.804
B 31 7 19 1,771
1956 c 29 494 633 453 1,874 7'266
D 49 550 442 9 1,817
Position
Totals 7,051 7,959 8,228 7,212 30,450 30,450

433,950.31 63

II

Total SS for SxPxY = 15,789,320 - C.T.
3

Spout Design SS (as before) = 17,515.26 3
Position SS (as before) = 20,303.85 3
SxP SS (as before) = 67,667.37 9
Years SS (as before) = 115,601.16 3
SxY SS (as before) = 23,411.72 9
PxY SS (as before) = 43,375.61 9

SxPxY SS (by subtraction) = 146,075.34 27

 

 

103
ANALYSIS III

DEPTH 0F TAPHOLE AND COMPASS POSITION
1956 SEASON

Yields in Poungs of Sugar

 

 

 

 

Depth of
Taphole North East South West Total
1.72 i.ޤ 3.33 1.86
9 o o . o o
2' 3.89 7.79 3.73 3.12 69°02
4.35 5.68 5.21 4.86
2.8; 3.22 %.g% 2.%8
3.2 . . 3. l
1*" 4.87 3.16 4.37 1.97 64‘1”
3.69 6.09 6.21 2.94
1.32 8.18 3.46 6.87
6" 3.76 2.78 14.93 3.61 70 11
2.94 7.13 2.10 3.64 °
1.52 HOE]. 80% 4.82
Analysis of Variance
Degrees Sum of Mean Value of
Source Freedom Sguares Sguare "F" Value F at
Total 47 173.12 -- -- —-
Position (P) 3 44.57 14.86 4.84** 2.86
Depth (D) 2 1.12 0.56 0.18 3.26
PxD 6 16.86 2.81 0.92 2.36
Error 36 110.57 3.07 -- --

 

104
.AJMLYSIS III (continued)
48 48 48

Total ss =21x2 - C.T. = 1,036.72 - 863.60 = 173.12
Position ss = 52:43 + (ZE)2 + (£5)2 + (DMZ - C.T.
12

10,891.9818 - . .
12 C T

908.17 - 863.60 = 44.57
(2’2")2 + (Zung2 + (26")2 _ C.T.
1

13,835.5534 _
1 C.T.

 

Depth SS

Two-Way Table for Interaction (PxD)

 

Dfiistfioig. 1426.4 228.2 22224 Teal 1461.1
2" 11.81 21.48 22.72 13.01 69.02
4" 14.67 16.96 21.64 11.20 64.47
6“ 9.59 22.60 18.91 1 .01 _1g,11
Total 36.07 61.04 63.27 43.22 203.60

C.T. = 863.60 (as before)
Total 33 for PxD =Zx'2 - C.T.

= 926.15 - 863.60 = 62055

Total SS for PXD - SSP - SSD
62.55 "' M057 "‘ 1.12 = 16.86

 

. .
I .
.
. . I I I a
n
. . .
. . . I .
_ .
o.
. I I
. I . . I I
I . . .
l .
u I I I I ..
. . .
I _
. I .
v . _
I . I
_ I I —
. o
I. u I
C I . I. . . '
I . . I I I i
n u _
I
_ . I
. u
. .
s
.
I 1
I I I l I I
I .. I
—
I O
, l

 

105
ANALYSIS III (continued)

12 12

$51166 = .7153

 

L.S.D..OS = 2.03 (.7153) = 1.45
L.S.D..Ol = 2.72 (07153) = 1095
S W N
5.27 5.09 3.60 3.01
1. 1.9 -

ANALYSIS IV

TAPHOLE DIAMETER AND COMPASS POSITION
(1953 80 1955 COMBINED)

Analysis of Variance

 

3°ch ' 33333;
Total 71
Year (Y) 1
Diameter (D) 2
YXD 2
Position (P) 3
YxP 3
DxP 6
YXDxP 6
Error 48

(1953) (1955)

Sum of
Sguares

362,643
114,561
16,093
2,076
14,619
5,847
16,061
20,006
173,380

Mean

Sguare

114,561
8.047
1,038
4,873
1,949
2,677
3,334
3,612

EX2 = 574.457 + 1.659.009 = 2,233,466

C.T. = (4,367 + 7,23922 = 134,699,236 = 1,870,823
7 7

=22x2 - C.T. = 2,233,466 - 1,870,823 = 362,643
(4,367)2 + (7.239)2 - C.T.
36

Total SS

Years SS

Diameter

Position

SS

SS

139859384 " 118702823 = 114,561
(3176h)2 + (4éi51)2 + (3,491)2 ‘ COT.

ll

1,886,916 - 1,870,823 = 16,093

"F"

Value

31o7**
2.2

1.3

106

"F"

4-04
3.19

 

2.80

(3,312)2 + (2.773)2 +8(2.899)2 + (2,622)2 - C.T.
1

1,885,442 - 1,870,823 = 14,619

 

 

107

ANALYSIS IV (continued)

Two-Way Table for Interaction (YXD)

 

 

 

Eggt 11161 111161 ;5[;§1 Year Total
1953 1,331 1,685 1,351 4,367
1955 2.43.3 2.6.6.6 2.140 .1232
D13??? 3.764 4. 351 3.491 11, 606
Total $3 for YxD = 2,003,553 - C.T. = 132,730 ‘2537
Year SS (as before) = 114,561 1
Diameter SS (as before) = 16,093 2
YxD SS (by subtraction) = 2,076 2

Two-Way Table for Interaction (YXP)
£333 Ngttg Egtt §QEEE Ettt Year Total
1953 V 1.232 933 1,185 1,017 4,367
1955 2,080 1,840 1,714 ;,ég5 _1,232
Pogifiiin 3,312 2,773 2,899 2,622 11,606
_D_,_F;,_
Total SS for YXP = 2,005,850 - C.T. = 135,027 7
Year SS (as before) = 114,561 1
Position SS (as before) = 14,619 3
YxP SS (by subtraction) = 5,847 3

108
ANALYSIS IV (continued)

Two-Way Table for Interaction (DxP)

Diameter North East South West Diameter Total

 

 

 

7/16" 1,015 1,000 922 827 3,764
11/16" 1,278 905 1,015 1,153 4,351
15/16" 1,019 868 962 642 3,491
Position
Total 3,312 2,773 2,899 2,622 11,606
D.F.
Total 33 for DxP = 1,917,596 - C.T. = 46,773 11
Position SS (as before) = 14,619 3
Diameter SS (as before) = 16,093 2
PxD SS (by subtraction) = 16,061 6

Three-Way Table for Interaction (DxPxY)

Diameter Year
Year Diameter North East South West Total Total

 

7/16" 408 28 53 286 1, 1
1953 137121: 44 381 £91 332 16% W
1955 1:;12: 888 53$ 22: 5%; 3,626 7 239
15/16" 665 587 581 307 2:140 ’

Position Total 3,312 2,773 2,899 2,622 11,606 11,606
D.F.

Total SS for DxPxY = 2,060,086 - C.T. = 189,263 23
Diameter SS (as before) = 16,093 2
Position SS (as before) = 14,619 3
DxP SS (as before) = 16,061 6
Year SS (as before) = 114,561 1
DxY SS (as before) = 2,076 2
PxY SS (as before) = 5,847 3
PxYXD SS (by subtraction) = 20,006 6

Source

Total

Year
Diameter (D)
YXD
Position (P)
YxP

DxP

YxDxP

Error

(195
52x2 = 1,912,

109
ANALYSIS V

TAPHOLE DIAMETER AND COMPASS POSITION

4
1

(1954 & 1956 COMBINED)

Analysis of Variance

Degrees Sum of Mean "F" "F"
Freedqg Sguares Sguare Value at 5%

71 727,118

1 2,926 2,926
2 47,648 23,824 2.77 3.19
2 183,483 91,742 10.66% 3.19
3 24,064 8,021
3 6,953 2,318
6 28,505 4,751
6 20,399 3,400

48 413,140 5,607

) (1956)
60 + 1,435,863 = 3,348,023

C.T. = (7,098 + 6,639)2 = 188,705,169 = 2,620,905
72 72

Total ss =§Zx2 - C.T. = 3,348,023 - 2,620,905 = 727,118

Year 33 = (7,098)2 36(6’639)2 - C.T.

= 2,623,831 - 2,620,905 = 2,926

Diameter SS

Position SS

(3,720)2 + (4,373)2 + (5,144)2 - C.T.

236689553 ' 236209905 = 47.643
(3,089)2 + (3,566)2 $80,171)2 + (3,911)2 - C.T.

2,644,969 ‘ 2,620,905 = 24,064

 

I
I
l
n.
a.
. c 1
..
2
I
I I
o
O I. u
. _
.
I C. I I
l D D I I l
. I p
I o
I I ‘
n.
. l u I I I n .
I I
n.
n _
0 I
o
n.
I
v
I I .
I

 

,1_—_—

 

llO
ANALYSIS V (continued)

TwoAWay Table for Interaction (YXD)

 

‘tht 8" 11161 11716" Year Total
1954 1,110 2,731 3,257 7,098
1956 2610 21.1.42 119131 ...6 _3_96
Diameter

Total 3,720 4,873 5,144 13,737

.JlgEL

Total 33 = 2,854,962 - C.T. = 234,057 5
Year SS (as before) = 2,926 1
Diameter SS (as before) = 47,648 2
YXD SS (by subtraction) = 183,483 2

Two-Way Table for Interaction (YxP)

 

jggag Nggth gagt Sggth Etgt Year Total
1954 1,503 1,885 1,575 2,135 7,098
1956 1 86 1,681 1,596 1,776 _§,63g
Position

Total 3,089 3,566 3,171 3,911 13,737

DeF.

Total SS for YXP = 2,654,848 - C.T. = 33,943 7
Year SS (as before) = 2,926 1
Position SS (as before) = 24,064 3
YxP SS (by subtraction) = 6,953 3

 

ANALYSIS V (continued)

Two-Way Table for Interaction (DxP)

 

 

 

 

 

QIEEEEEE 42224 $222 §22£Q 4222 212$2E2£_22221
3/8 " 1,033 802 843 1,042 3,720
7/16" 875 1,321 1,189 1,488 4,873
11/16" 1,181 1,443 1,139 1,381 5,144
Position
Total 3,089 3,566 3,171 3,911 13,737
.21E1
Total $3 for DxP = 2,721,122 - C.T. = 100,217 11
Position SS (as before) = 24,064 3
Diameter SS (as before) = 47,648
PxD SS (by subtraction) = 28,505 6
Three-Way Table for Interaction (DngYl
Diameter Year
X232 Diameter 423th EEEE §22£E Ettt Total Tgtgl
1954 77123 5&8 781 611 871 2:73? 7,098
11/16" 624 993 750 890 3,257
' 3 8" 702 611 629 668 2,610
1956 7 16" 327 620 578 617 2,1 2 6,639
11/16" 557 450 389 491 1,8 7
Position Total 3,089 3,566 3,171 3,911 13,737 13,737
D.F.
Total ss for DxPxY = 2,934,883 - C.T. = 313,978 23
Diameter SS (as before) = 47,648 2
Position SS (as before) = 24,064 3
DxP SS (as before) ' = 28,505 6
Year SS (as before) = 2,926 l
DxY SS (as before) = 183,483 2
PxY ss (as before) = 6,953 3
PxYxD SS (by subtraction) = 20,399 6

 

 

1

112
ANALYSIS VI

TYPE OF CONTAINER
(1956 SEASON)

Yields in Pounds of Sap

 

 

 

 

Metal Plastic
North East ' South West North Egst South West
81.3 185.1 167.6 131.4 65.6 187.4 416.9 152.1
161.5 135.7 215.2 87.5 249.0 295.1 143.1 227.4
237.4 149.9 199.7 93.4 68.5 248.4 320.3 108.7
120.2 348.5 195.8 . 177.1 191.9 189.2 97.2
156.0 150.2 295.6 318.2 272.7 231.8
176.5 268.9 149.7 328.6 201.6
138.3 208.2 232.3
87.5 104.6
357.4

 

 

1158.7 1238.3 582.5 1161.6 560.2 1569.6 2238.1 817.2

Means:
144.8 206.4 194.2 165.9 140.1 261.6 248.7 163.4
Total SS = 2,139,117.2 - 1,812,041.80 = 327,075.44

Between Classes SS =E;QIE _ C.T.
n

(1158.7)2 + (1238.3)2 + . . . - <81 -2 2 - . .
8 6 ___%_.1_ c T

1,908,441.26 - 1,812,041.80 = 96,399.46

 

 

 

 

 

 

113
ANALYSIS VI (continued)
Analysis of Variance
for error)
Degrees Sum of Mean "F" "F"
Source Freedom Sguares Sguare Value at 5%
Total 47 327,075.44
Between 7 96,399.46 13,771.35 2.39% 2.25
Within 40 230,675.98 5,766.90
Adjustment of Class Numbers
Metal Plastic
Position Observed Expected Observed Expected Total
No. No. No. No.
North 8 6 4 6 12
East 6 6 6 6 12
South 3 6 9 6 12
West _1 6 _5 6 42
Total 24 24 48

Expected N0. = Row Total

Grand Total x 001 Total

= fig x 24 = 6

The Chi Square test of expected numbers (for interaction)

Chiz‘=:£:(0bserved No. - Expected No.)2
Expected N0.

= 4+9+l+4+9+1=28=4.66
7?? 27
Chi2 for significance at 5% for 3 D.F. = 7.8

We can therefore assume negligible interaction and
proceed to adjust totals by expected numbers.

 

0 .1
. 1.0.5.6 .7

.. ’PP‘IWm..nE.I..I1.*"W‘nJl-JA ,
II \IIII r 0..”
v

8|} 1|!
GI"! .1.

114
ANALYSIS VI (continued)

Two-Way Table for Expected Totals

 

Metal Plastic
Exp. Observ. Exp. Exp. Observ. Exp. No.
No. Average Total No. Average Total Tot. X

 

6 144.84 869.04 6 140.05 840.30 12 1709.34
6 206.38 1238.28 6 261.60 1569.60 12 2807.88
6 194.17 1165.02 6 248.68 1492.08 12 2657.10
6 6

165.94 995.64 163.44 980.64 44 1976.28
24 4267.98 24 4882.62 48 9150.60

Sit/313712

Total SS =:(Exnected Tot.)2 - (XExpected Tot.)‘2
6 71.8

= l,832’u7h.0h - 197%,)411-7951 = 88,026.53
Position 88 =§;§E - C.T. = 21,771,896.38 - C.T. = 69,877.19
12 12

Container ss =5:02 - C.T. = o 6 1. - C.T. = 7,870.46
24‘ 2

Analysis of Variance for Totals

Source Degrees Freedom Sum of Sguares Mean Sguare
Total 88,026.53 --

7.870.46 7,870.46
10,278.88 3,426.29

7

Position, 3 69,877.19 23,292.40
Container 1
3

PxC

 

ANALYSIS VI (continued)

2211222
Position
Container
PxC

Error

Analysis of Variance of Adiusted Values

Degrees
Freedom

3
1
3
40

Sum of

Sguares

69,877.19
7,870.46
10,278.88
230,675.98

Mean
Sguare

23,292.40
7,870.46
3,426.29
5,766.90

"F 1!

Value

4-O4*
1°36

115

"F I!

at 5%
2.84
4.08

 

ANALYSIS VII
COMPASS POSITION

(COMBINED DATA FROM 3 EXPERIMENTS)

1956 SEASON

Combined Yields in Pounds of Sap

 

 

 

Repl. (Eiper. (Eiper. (Eiper. Totals
#2) #3) #4)
N 1,336.5 1,718.9 1,586.8 4,642.2
E 1,887.5 2,807.9 1,680.4 6,375.8
S. 2,200.1 2,820.6 1,595.7 6,616.4
W 1,842.7 1,978.8 1,776.4 5,592.9
Totals 7,266.8 9,326.2 6,639.3 23,232.3
Analysis of Variance
Degrees Sum of Mean "F"
Sgtggt Freedom Sguares Sguare YEAES
Total 131 800,670.35
Repl. 2 47,690.06 23,845.03 4.482%
Position 3 72,095.12 24,031.71 4.52**
RxP 6 42,938.63 7,156.44 1.35
Error 120 637,946.54 5,316.22

116

Mean
Yields

140.7
193.2
200.5
169.6

"F I!

at 1%

4.78
3.94

 

 

 

117
ANALYSIS VII (continued)

C.T. - 021:)2 = (23232.3)2 = 4,088,937.60
132 132

Total SS =§:X2 (for all three experiments) - C.T.
= 4.889.607-95 - 4,088,937-60 = 800,670.35
88;. = (2N)2 + (XE)2 + (ZS)2 + (DMZ - C.T.
33

4,161,032.72 - 4,088,937.60 = 72,095.12

SSRepl. = (RI)2 + (Re)2 + (R )2 - C.T.
48 36

 

= 2,912,174.76 + 1,224,452.90 - 4,088,937.60 = 47,690.06

Total 83 for Interaction =Z(Xl)2 +X(X2)2 +3339)2 - C.T.
12 9

 

= 3,024,602.10 + 1,227,059.31 - 4,088,937.60 = 162,723.81

SSPxR = Total SS for Interaction - SSP - SSR

= 162,723.81 - 72,095.12 - 47,690.06 = 42,938.63

67.?1 " fij) = ”2 x Ergé Mo ‘90 = 1(10163§§.QA
V322.19515 = 17.95

L.S¢Do.05 = 17.95 x 1.98 = 350514.
L.S.Do.01 = 17095 x 2062 = (4.7003
S E W N
200.50 193.21 169.63 140.67
169.63 140.62 140062

30.87 52.54** 28.96

I
4
. C
a
7 . I I u
I
I o
I I
_
I
_
a
.. I I I I I
I
u o
t . I
I
I. I
a
I I C C
. .
I
I
I.
1
i .
I
I
I
o
o
u
C

 

118
ANALYSIS VIII

NUMBER OF BUCKETS PER TREE
(FOUR YEARS DATA COMBINED)

Computation of Average Diameter

 

 

 

 

 

 

One-Bucket Trees Two-Bucket Trees
Tree No. DBH Basal Area Tree No. DBH Bgsal Area
3 16.3 1.4491 12 17.3 1.6324
9 16.0 1.3963 13 17.1 1.5948
25 16.1 1.4138 18 16.9 1.5578
32 16.0 1.3963 50 16.2 1.4314
92 17.6 1.6895 ' 61 16.6 1.5029
97 16.3 1.4491 63 17.5 1.6703
98 16.8 1.5394 84 16.4 1.4669
102 16.4 1.4461 109 17.0 1.5763
119 16.7 1.5211 » 110 17.2 1.6136
Total Basal Area 13.3015 Total Basal Area 14.0464
Aver. Basal Area 1.4779 Aver. Basal Area 1.5607
Aver. Diameter 16.463 Aver. Diameter 16.916
Two-Bucket Trees Three-Bucket Trees
Tree No. DBH Basal Area Tree No. DBH Basal Area
10 21.4 2.4978 41 20.0 2.1817
38 19.8 2.1382 42 21.8 2.5920
60 20.6 2.3145 99 22.4 2.7367
82 19.2 2.0106 100 20.3 2.2476
83 19.3 2.0316 111 21.1 2.4282
112 19.5 2.0739 115 21.9 2.6159
113 19.2 2.0106 116 20.1 2.2035
123 19.3 2.0316 118 20.6 2.3145
Total Basal Area 17.1088 Total Basal Area 19.3201
Aver. Basal Area 2.1386 Aver. Basal Area 2.4152

Aver. Diameter 19.802 Aver. Diameter 21.042

 

 

 

119
ANALYSIS VIII (continued)

Yields in Pounds of Sap

1253 1254 1255 19 6 Totals

 

145.6 283.6 152.1 201.6
140.8 81.9 255.3 295.6
123.5 150.1 307.9 177.1
Single_ 199.9 383.5 392.4 295 .1
Bucket ' 199.1 326.7 215.6 328.6
Trees 238.1 246.0 381.2 353.2
255.8 375.8 302.9 215.2
182.8 328.4 373.4 357.4
199.2 311.7 319.4 268.9 9,565.6
230.7 277.4 174.0 122.6
267.8 243.1 254.5 140.7
128.0 237.8 208.8 183.3
271.0 376.6 284.7 296.6
Two-
Bucket 139.4 358.2 392.8 301.2
Trees 198.5 215.7 258.2 215.8
153.2 296.5 276.6 260.9
293.2 222.3 413.8 470.7
290.6 470.9 291.5 420.1 9,637.7 ‘

 

 

 

Tetals 3.657-4 5338602 5,255.1- 4,904.6 19320303
Total 88 = 5,722,664.23 - (19,203.3)2
72

= 5,722,664.23 - 5,121,760.15 = 600,904.08
Number 88 = (9,565.6)2 +6(9,637.7)2 - C.T.
3 .

= 5,121,832.35 - 5,121,760.15 = 72.20
Year 88 =(3,657.4)2 +(5,386.2)2 +8(5,255.1)2 +(4,904.6)2 - C.T.
1

= 5,225,494.58 - 5,121,760.15 = 103,734.43

 

 

120

ANALYSIS VIII (continued)

Two-Way Table for Interaction (YXN)
Number

1251 1254 .1255 l9 6 221212
Single-Bucket 1,685.0 2,687.7 2,700.2 2,492.7 9,565.6
Two-Bucket 1,972.4 2,698.5 2,554.9 2,411.9 9,637,7
Year Totals 3,657.4 5,386.2 5,255.1 4,904.6 19,203.3

Total SS for Interaction

LEE 2084262902; " C.T.
9

 

 

 

 

. D.F.
= 109,865.32 7
Year SS (as before) = 103,734.43 3
Number SS (as before) = 72.20 1
YxN SS (by subtraction) = 6,058.69 3
Agglysis of Variance
Degrees Sum of Mean "F" "F"
Source Freedom Sguares Sguare 23433 gt_;%
Total 71 600,904.08
Years 3 103,734.43 34,578.14 4.51** 4.10
No. Holes 1 72.20 72.20
YxN 3 6,058.69 2,019.56

Error 64 491,038.76 7,672.48

 

 

ANALYSIS VIII (continued)

Two-
Bucket
Trees

Three-
Bucket
Trees

Yields in Pounds of Sap

121

 

 

122; 1254 1222 19 6 222215

340.6 566.7 370.1 361.9

364.5 761.2 566.5 396.1

264.8 493.5 425.9 393.8

158.5 307.2 195.2 284.2

323.1 474.3 559.8 424.1

252.2 353.2 467.1 551.4

163.6 210.7 335.4 272.3

317.8 470.1 248.4 310.5 11,986.0

491.7 1,084.1 797.8 390.7

400.1 442.7 461.0 393.3

283.6 628.6 553.8 415.4

212.6 439.5 281.1 492.7

254.9 304.5 529.7 528.0

214.9 306.3 355.6 411.1

246.3 601.5 486.6 297.4

270.1 519.3 515.0 533.5 14140.4 '
4,559.3 7,960.4 7,149.0 6,457.7 26,126.4

Total ss = 12,307,75o.46 - (26,126.4)2 = 1,642,300.82

= (11,986.0)2 + (14,140.4)2 - C.T.
32

Number S

Year SS

S

= 10,737,972.13 - 10,665,449.64 = 72,522.49
(4,559.3? +(7,960.4)2 +(7,149.0)2 +(6.457.7)2 - c.T.
16

11,060,329068 " 10,6659M906LI. = 391.1,880004

 

ANALYSIS VIII (continued)

Two-Way Table for Interaction (NxY)

122

 

 

 

 

 

, ‘ Number
1953 1954 1955 19 6 Totals
Two- ‘
Bucket 2,185.1 3,636.9 3,168.’ 2,995.6 11,986.0
Three- ‘ ‘
Bucket 2 3 .2 32 . 3,980.6 3,462.1 14,140.4
Totals 4,559.3 7,960.4 7,149.0 6,457.7 26,126.4
' D.F.
Total SS = 8921‘4E8(2020 " C.T. = 481,409.38 7
Number SS (as before) = 72,522.44 1
Year SS (as before) = 394,880.04
NxY SS (by subtraction) = 14,006.85 3
Analysis of Variance
Degrees Sum of Mean "F" "F"
Source Freedom Sguares Sguare Value at 0
Total 63 1,642,300.82
Year 3 394,880.04 131,626.68 6.35** 2.78
No. Holes 1 72,522.49 72,522.49 3.50 4.02
YXN 3 14,006.85 4,668.95
Error 56 1,160,891.44 20,730.20

 

 

 

 

r—‘—’—i —. 1, ,,

ROOM USE ONLY

 

ROOM USE QMLYWW; .,

W7

 

 

 

 

”I11111111441111ES