EFFECTS. OF LIGHT: DARK CYCLES
0N DEVELOPMENT IN
DRDSOPHILA MELANOGASTER

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
DALE L. CLAYTON
1968

  

 

Wt” WMflWm/I/flfl/W/ _ L I B R A :j
2 6250 Michigan State
University

This is to certify that the

thesis entitled

Effects of light-dark cycles
‘ on development in
Drosophila melanogaster

presented by

Dale L. Clayton

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in M

 

// '
Date // 4767/4]? /?é78

 

 

 

ABSTRACT

EFFECTS OF LIGHT:DARK CYCLES
ON DEVELOPMENT IN
DROSOPHILA MELANOGASTER

 

by Dale L. Clayton

Body of Abstract

A functional role for circadian rhythms in developmental pro-
cesses has not been described heretofore. This thesis tests the
hypothesis that exogenous light cycles act as a chronometer in regu-
lating develoPmental processes, i.e. longer circadian light cycles
retard developmental rate.

One hundred twenty Drosophila melanogaster eggs were placed in
each of 40 culture vials. These vials were placed into 6 treatment
groups. The treatment groups were 1) long cycle (25 hrs. 9 min.)
with 15 second twilight transitions, 2) long cycle with 15 minute
twilight transitions, 3) short cycle (22 hrs. 52 min.) with 15
second twilight transitions, 4) short cycle with 15 minute twilight
transitions, 5) constant light and 6) constant darkness. Long and
short cycle groups were in phase when flies began to eclose at
251.5 hours of incubation, and were continued in constant light
thereafter.

Time from oviposition to eclosion was used as a measure of

developmental rate. Flies incubated under short cycles eclosed
earlier (P‘:0.01) than flies incubated under long cycles. Flies
incubated under 15 minute twilight transitions eclosed earlier
(P‘:0.01) than flies incubated under 15 second twilight transi-
tions. Flies incubated under constant light eclosed earlier
(P‘<0.0l) than flies incubated under constant darkness.

Mortality differed significantly (P<r0.0l) between flies
incubated in constant or in cyclic conditions. Mortality was
higher in constant conditions. This difference precludes com-
parison of eclosion rates between these conditions since eclosion
and mortality are probably not independent. Differences in mor-
tality were insignificant (P4<0.01) in all other comparisons.

The conclusions are that circadian light cycles do act as a
chronometer for developmental processes as indicated by time of
eclosion; and furthermore, circadian light cycles increase develop-

mental success as indicated by decreased mortality.

EFFECTS OF LIGHT:DARK CYCLES
ON DEVELOPMENT IN

DROSOPHILA MELANOGASTER

By

Dale L. Clayton

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Zoology

1968

’r 3"" Fry-1.
,3 a / A: ,7

A)- "21/." 6:?

ACKNOWLEDGMENTS

I am indebted to:

My committee, Dr. M. Balaban, Chairman; Drs. Wm. Cooper,
G. I. Hatton, and J. 1. Johnson for valuable criticism
of this thesis and for meeting somewhat unreasonable
demands for their time.

Kay, for long hours of typing and retyping.

Carola Cattani,for making her lab and constant temper-
ature room available.

ii

TABLE OF CONTENTS

Acknowledgments . . ... . . . . . . . . . . . .
List of Tables . . . . . . . . . . . . . . .
List of Figures. .

List of Appendices . . . . . . . . . . . . . . .
Introduction .

Methods and Materials.

Results: 0 O D O O O O I O O O O O O I O O

I. Effects of light cycles and twilight transitions

on mortality. . . . . . . . . . . . . . . . .

II. Effects of light cycles on eclosion . . . . .

III. Effects of twilight transitions on eclosion .
IV. Effects of sex on eclosion.

Discussion . . . . . . . . . . . . . . . . . . . . . .

I. Effects of light cycles and twilight transitions

on mortality. 0 O O O O O O O O I O I O O O 0
II. Effects of light cycles on eclosion . .
III. Effects of twilight transitions on eclosion .

IV. Effects of sex on eclosion. . . . . . . .

werView O O 0 O O I O O O O 0 O O O O I O O O O 0 O 0

Summary. . . . . . . . . . . . . . . . . . . .

Bibliography . . . . . . . . . . . . . . . . . . .

iii

page
11

iv

. vii

ll
ll
l3
17

17

22

22
24
25
28
3O
33

41

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

10.

ll.

12.

13.

LIST OF TABLES

Experimental Design .
Summary of x2 mortality statistics.

Comparison of mortality for all treatment
groups.

Summary of data comparing mortality rates
of all six treatment groups .

Summary of X2 eclosion rate statistics.

Summary of data used for comparing eclosion
rates of long and short cycle groups.

Summary of data comparing eclosion rates of
long and short cycle treatment groups .

Raw data for the long cycle, 15 second twilight
treatment group . . . . .

Raw data for the long cycle, 15 minute twilight
treatment group . . . . . . . . . . . . . .

Raw data for the short cycle, 15 minute twilight
treatment group . . . . . . . . .

Raw data for the short cycle, 15 minute twilight
treatment group .

Raw data for the constant dark treatment group.

Raw data for the constant light treatment group .

iv

page
10

12

12

13

16

16

16

35

35

36

36

37

37

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

10.

11.

LIST OF FIGURES

Motor driven Light:Dark boxes.

Comparison of the long and short cycle
light regimes.

Eclosion curves of long and short cycle
lightzdark treatment groups, using a 15
second twilight transition (males and
females combined). .

Eclosion curves of long and short cycle
1ight:dark treatment groups, using a 15
minute twilight transition (males and
females combined).

Eclosion curves of constant light and
constant darkness treatment groups
(males and females combined)

Eclosion curves of long and short cycle
lightzdark treatment groups, using a 15

second twilight transition (males only).

Eclosion curves of long and short cycle
1ight:dark treatment groups, using a 15

minute twilight transition (males only).

Eclosion curves of long and short cycle
light:dark treatment groups, using a 15

minute twilight transition (males only).

Eclosion curves of long and short cycle
1ight:dark treatment groups, using a 15

minute twilight transition (females only).

Eclosion curves of constant light and
constant darkness treatment groups
(males only)

Eclosion curves of constant light and
constant darkness treatment groups
(females only)

page

, 14

, l4

, 15

. l9

. 19

.20

.20

.21

.21

Figure 12.

Figure 13.

Figure 14.

Schematic representation of the twilight
transitions used . . . . . . . . . . . . . . . . . .27

Density dependence of eclosion rate for
flies incubated in short lightzdark
cycles (22 hr. 52 min.). . . . . . . . . . . . . . .40

Density dependence of eclosion rate for
flies incubated in long Iight:dark
cycles (25 hr. 9 min.) . . . . . . . . . . . . . . .40

vi

LIST OF APPENDICES

Page
Appendix I. Raw data . . . . . . . . . . . . . . . . . . . . . 34
Appendix II. Density dependent development rates. . . . . . . . 38

vii

INTRODUCTION

The existence of circadian rhythms* in developing organisms is
well documented (Rutenfranz 1961, Petren and Sollberger 1953, Johnson
1966, Hellbrugge.g£,.gl: 1964). However, a functional role for these
rhythms in the developmental process has not been described. Harker
(1965a) suggests that competence of embryonic tissues cycles on a
circadian basis, but does not describe the functional significance
of this. The abundant evidence of light cycles regimenting physio-
logical and behavioral events leads one to suspect they play a role
in the chronology of developmental events. Demerec (1950) shows
that in Drosophila there is a fixed sequence of developmental events
leading from one stage of deve10pment to the next. Assuming that the
chronology of these events is affected by external environmental
oscillations, developmental time should be lengthened or shortened
by lengthening or shortening the daily light cycles.

The hypothesis this thesis tests is:

The developmental period, from ovisposition to
eclosion of the Drosophila melanogaster imago,
is longer when flies are incubated in long
circadian light:dark cycles, than when they are
incubated in short circadian lightzdark cycles.

This hypothesis implies that the length of the developmental

period is correlated with the number of light:dark cycles experienced.

Harker (1965a and 1965b) recognizes this possibility but presents

*Circa=about; dian=day, i.e. oscillations of about one day.

1

data aimed at disproving it. I find no fault with her data but do
find them inadequate for the conclusions she draws.

Harker's data are significant in that she points to an important
relationship between developmental time Spans and the entrainment to
daily light:dark cycles. She finds that time taken to complete three
stages of Drosophila pupal development is relative to the phase of
the light cycle at which each stage is entered. For example, the
number of hours required between head eversion and the appearance of
yellow eye pigmentation was 19 to 20 hours if the head everted four
hours after the lights came on, and 54 to 55 hours if the head everted
one hour after the lights went off. These data present a powerful
demonstration that circadian light cycles affect development.

She also shows that pupae kept in darkness, once a stage is
entered, develop with the same time span as pupae maintained in 24-
hour light:dark cycles. Harker (1965a, p. 329 and 327) states, "It
is clear from the results that the same developmental periods are
involved whether the pupae are in a light cycle or in constant dark-
ness." She further concludes, "There does not seem to be any corre-
lation between the length of the developmental period and the number
of bright light:dim light cycles, or any sequence of these cycles,

. . . " This conclusion follows only if the light cycles are restric-
ted to a period of 24 hours or very nearly 24 hours. Harker (1965a
and 1965b) used only 24-hour 1ight:dark cycles. The free-running

period* of Drosophila is very close to 24 hours. Therefore, one

 

*One free-running period equals the time required for one oscillation
of the internal clock under constant environmental conditions, i.e.,
the period is not entrained to environmental oscillations but is
"running free".

would not expect develOpmental time to differ between one group
entrained to a 24—hour light cycle and another group following a
24-hour free-running cycle since the internal clock would maintain
the same phase relations in both groups. Harker does not present
sufficient data to arrive at a conclusion concerning cycles longer
or shorter than 24 hours. In terms of 24-hour cycles, her conclusion
is not meaningful because the endogenous cycle continues on a 24-hour
free-running basis whether the exogenous cycle is Suspended or not.
Data given by Shutze gt: a}, (1962) and by Minis and Pittendrigh
(1968) can be used to argue that there may be a correlation between
the number of cycles of the internal physiological clock and develop-

mental rates in chickens and in the moth, Pectingphora gossypiella.

 

Shutze et. al. found that chicks (diurnal animals) incubated in
constant light hatched approximately 16 hours before chicks incubated
in constant darkness. Assuming that Aschoff's Rule* is operating in
the ontogeny of chick embryos, we would expect that oscillations of
the internal clock would occur more rapidly for chicks incubated in
constant light.

If it is true that a discrete set of developmental events occur
with each oscillation of the internal clock, then knowing the number
of oscillations that the internal clock has completed should be use-
ful for predicting stages of development (e.g. hatching). Twenty-one
*Aschoff's Rule (Aschoff 1958 and Pittendrigh 1960) states that
organisms maintain relatively constant free-running circadian
periods when subjected to constant conditions; and~that this
period is shorter in constant light than in constant darkness

for diurnal organisms, but longer in constant light for nocturnal
organisms. This principle was first described by Johnson (1939).

days of incubation (i.e. 21 x 24 hours), normally associated with
hatching in chickens, would then be an artifact of the natural
Vivarium.

Minis and Pittendrigh (1968) find that the median time required
for embryogenesis of the moth, g, gossypiella at 20°C. (i.e. oviposi-
tion to hatching from the egg) is 244 hours in constant light and
275 hours in constant darkness. If we can think of the moth as being
diurnal during embryogenesis, Aschoff's Rule can be applied to these
data as it was to the data of Shutze gt; 5;; (1962) for the chick
embryo. The adult moth is nocturnal. However, hatching is keyed to
the light portion of the light:dark cycle (Minis and Pittendrigh 1968)
as is hatching and emergence of Drosophila, a diurnal insect. This
suggests that the moth embryo functions on a diurnal basis.

Circadian organization is present in this moth for at least the
last half (5% days) of embryogenesis. This is based on entrainment
of hatching to a single light perturbation. During the first 5% days
of embryogenesis, circadian organization is not evident nor is there
any apparent difference in hatching times between embryos incubated
in either constant light or constant darkness for the first 5% days
and maintained in similar conditions thereafter. During the second
half of embryonic develOpment, however, both differential developmental
rate (in response to light intensity) and circadian organization are
present. This adds credence to the concept of a functional tie-up be-
tween developmental rate and circadian organization.

If we accept the assumption that Aschoff's Rule is operating
during embryogenesis, the data of Shutze et. al. (1962) and of Minis

and Pittendrigh (1968) support the hypothesis that there is a

correlation between the length of the developmental period and the
number of circadian oscillations.

The possibility of light intensities acting on developmental
rate (via Aschoff's Rule) has been suggested by Thomas and Pizzarello
(1967). Their study employed blind and sighted girls as subjects and
used age of first menarche as a measure of sexual maturation. Since
blind girls experience constant darkness, they were expected to
exhibit longer free-running periods and therefore slower development
than sighted girls. Thomas and Pizzarello found no difference between
blind and sighted girls in time of first menarche.

All subjects were from a school for the blind or from a children's
home. Institutional life is highly regimented and both sets of girls
followed a rigid 24-hour schedule. This schedule differs from constant
conditions assumed by Aschoff's Rule and the prediction would logically
be "no difference".

I propose to use light cycles that are either longer or shorter
than 24 hours. This allows specification of cycle lengths, and
assumptions based on Aschoff's Rule are unnecessary. The endogenous
rhythms of Drosophila are sensitive to resetting by light during early
stages of development. Brett (1954, p. 176) states that the rhythmic
entrainment of emergence ”was shown to persist under conditions of
constant darkness and temperature provided that the animals [Drosophila]
had previously been subjected to day:night illumination changes at
some developmental stage other than prelarval." The prelarval period
ends approximately 24 hours after oviposition. This means that the
endogenous cycle is sensitive to resetting by day:night illumination

changes for the last 9 or 10 days of incubation.

METHODS AND MATERIALS

Adult Drosophila melanogaster oviposited eggs on a thin layer of
agar media for four hours (7:00 a.m. to 11:00 a.m. EDST). Eleven a.m.
was designated as the start of incubation. Small blocks of agar
containing 120 eggs* were transferred to glass vials (3 inches long
and 1 inch in diameter) into which approximately 8 ml. of media had
been poured. These vials were plugged with cotton, divided randomly
into six treatment groups and placed in six Light:Dark boxes as illus-
trated in Figure 1. These boxes were made of 1/4" composition board
and painted silver inside and out. A beveled strip of wood was glued
down the center of the box. The base of the vials rested on this
strip and the t0ps were tilted against the side of the box.

A black construction paper shutter was glued to a 1/4" wooden
dowel. The dowel protruded through 1/4" holes centered near the top
of either end of the box. When the dowel shaft was rotated until the
shutter was vertical, light entered the box freely; when the shutter
was horizontal, light could not enter the box. This provided the
conditions of light and dark respectively. Thin,black polyethylene
was fastened to, and overlapped, the construction paper shutter to
form a light-tight seal.

One-round-per-hour synchronous motors were coupled to the shutter

shafts of two of the boxes. One-round-per-minute synchronous motors

*The number of eggs per vial was held constant because eclosion rates
are density dependent (i.e. eclosion is slower in vials with higher
densities). See Appendix II.

were coupled to the shafts of another two boxes. The remaining two

boxes were Operated manually.

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. Motor driven Light:Dark boxes. At the time
preset on the control clock (cc) the shutter turns 900
simulating either dawn or dusk. Limiting switches (ls)
check the travel of the shutter. The clock and the
light-dark switches must be reset before the following
twilight transition will occur. Each clock controls
long or short cycles for shutters driven by one-RPM and
one-RPH synchronous motors. '

A time clock was set to trigger the onset of each twilight
transition (i.e. light-dark or dark-light) and limiting switches
checked the travel of the motor-driven shutters. Switches on the
front of the boxes were set to indicate twilight transition desired

(i.e. dawn, dark to light; or dusk, light to dark). Light and dark

phases of each cycle were of equal duration.

Each twilight transition required a 900 turn of the motor~driven
shutter. This represents a 15 second twilight transition for the 1
RPM motor (i.e. 900/360O x 1 RPM = 15 sec.) and a 15 minute twilight
transition for the 1 RPH motor (i.e. 900/360o x l RPH = 15 min.)

The six boxes were set up as 3 modules of 2 boxes each. One
time clock was set for a short circadian cycle (22 hr. 52 min.) and
controlled the shutters of 2 boxes, one with a 15 second twilight
transition and the other with a 15 minute twilight transition. A
second time clock was set for a long circadian cycle (25 hr. 9 min.)
and controlled the shutters of 2 boxes, one with a 15 second twilight
transition and the other with a 15 minute twilight transition. The
remaining two boxes were set manually; one for constant darkness, the
other for constant light.

These particular long and short cycle lengths were chosen so
that the cycles would be in phase with each other at 251.5 hours when
the flies began to eclose. At this time the short cycle boxes had
completed exactly one more light cycle than the long cycle boxes (see
Figure 2).

Eclosion is strongly keyed to dawn, therefore light cycles not
in synchrony cause differences in eclosion times. If the long and
short cycles were not in phase at the onset of eclosion, it could be
argued that the cycles being out of phase were reSponsible for any
difference in eclosion time of the treatment groups. It may also be
argued that cycles longer or shorter than 24 hours are constantly
being reset, and that the endogenous cycle lags or leads the light
cycle. This is probably the case. Inspection of Figure 2 will show

that the next to the last dawn preceeded the final dawn by 25 hours

9 minutes for the long cycle group and 22 hours 52 minutes for the
short cycle group. If eclosion were expected 24 hours* after the
next to the last dawn, it would occur 2 hours 17 minutes earlier for
the long cycle group. This would not support the hypothesis of this
thesis (page 1). Therefore this bias can not be used to argue

against data supporting this hypothesis.

 

 

 

 

 

 

 

A B c
1-.-.-.1 1-.'.°.1 1.221 1'.°.'.1 12.-.1331 1-.'.-.1 T‘vv-I 1-.-.-.1 1.-.-.

flat] 1-.-.-1 1.-:.1 1.31 1.-.-.1-.'11.-.-.1 1-.~.-1 13.-.1 1:1 1.-..

0 hrs. 126.75 hrs. 251.5 hrs.

Figure 2. Comparison of the long and short cycle light
regimes. The top line represents the long cycle (25 hr.
9 min.) and the bottom line the short cycle (22 hr. 52
min.). Point A represents the onset of incubation; point
B, the time at which the treatment groups were 1800 out of
phase and Point C, the time at which the long and short
cycles differ by exactly one 1ight:dark cycle. Point C
also represents 1) the onset of constant light, and 2)
the onset of eclosion. The first flies were collected

at 253 hours of incubation and those eclosing thereafter
at 4 hour intervals. Stippled areas indicate the dark
phase of the light:dark cycle.

After 251.5 hours of incubation, vials of all treatment groups
were subjected to continuous light. Starting at 253 hours of incu-
bation, flies that had eclosed in each vial were removed and counted
at four-hour intervals.

Room conditions of light and temperature were held constant at
15 foot candles and 23.50C. The number of eggs starting incubation
*24 hours is the natural (i.e. free running) period of the endogenous

cycle. This is particularly true of pOpulation cycles which are the
averages of several near 24-hour individual periods.

10

and the number of adults successfully completing incubation were
recorded for each treatment.

The constant condition groups were included in this design to
provide baseline data. Intuitively, a group subjected to a 24-hour
cycle (i.e., intermediate to the long and short cycles) may seem a
more apprOpriate control group. This is not realistic, however. If
incubation (i.e. cycle one) is begun at 1100, the lights come on to
begin the 11th long cycle or the 12th short cycle at 2230. This is
just 30 minutes before the lights go out to mark 11%, 24-hour cycles.
The control group would then be 180° out of phase and could not
provide appropriate baseline data.

Table 1 illustrates the experimental design. I shall deal with
the following aspects of this design:

I. Effects of light cycles and twilight transitions on mortality.
II. Effects of light cycles on eclosion.
III. Effects of twilight transitions on eclosion

IV. Effects of sex on eclosion.

Table 1. Experimental Design

 

. . _ 15 minute twilight transition*
Long Cycle (25 hr. 9 mini)
15 second twilight transition*

15 minute twilight transition*
Short Cycle (22 hr. 52 min.)
15 second twilight transition*

Constant light*
Constant Conditions
Constant darkness*

 

*Males and females were recorded for each treatment group.

RESULTS

Differences in eclosion rates and mortalities were tested with
Chi square contingency tests. A significance level of 0C = 0.01 was
chosen. The eclosion curves of the 1ight:dark treatment groups are
bimodal and therefore not suitable for testing with parametric tests
that assume a normal distribution.

I. Effects of light cycles and twilight transitions on mortaligy.
Differences in mortality rates were tested by comparing the number of
flies eclosing and not eclosing. This comparison is based on a known
number of eggs (120 per vial) at the start of incubation.

Four independent orthogonal comparisons of mortality can be made
of the light:dark and the constant darkness treatment groups. The
comparisons chosen were 1) long cycle versus short cycle, 15 second
twilight, 2) long cycle versus short cycle, 15 minute twilight, 3)
fifteen second twilight versus 15 minute twilight, and 4) cyclic
conditions versus constant darkness. These comparisons and their
associated ){2 values are listed in Table 2.

Of these, only the cyclic conditions and constant darkness
comparison showed a significant difference in mortality (P< 0.01).
Mortality was greater in constant darkness. Since mortality was 2.3%
higher in constant light than in constant darkness, it can be inferred
that mortality of flies incubated in constant light is also signifi-
cantly higher than mortality of flies incubated in cyclic conditions.

A fifth independent comparison of mortalities was made between

11

12

constant darkness and constant light conditions. This comparison
shows no significant difference (Table 2). The constant light treat-
ment group was excluded from the orthogonal contrasts to allow direct
comparison of the constant darkness group with the cyclic groups.
Percent mortality for each treatment group is given in Table 3.

Table 4 is a summary of the data used for calculating x2 and percent-

age values of mortality.

Table 2. Summary of x2 mortality statistics

Treatment 2 P .Relative
Groups 3L Value Mortality

 

 

Long versus short/15 second

twilight transition 7x?(1) 1.298 >0.20 No difference

Long versus short/15 minute
twilight transition '2?(1) = 1.448 >0.20 No difference

15 second versus 15 minute
twilight transition ‘y¥(1)

3.428 >0.05 No difference

Cyclic versus
constant conditions

2 _ Cyclic ( Con-
2 (1) 9'048 (0'005 stant‘ darkness

Light versus dark
constant conditions 78(1) - 1.238 )0.20 No difference

 

 

 

 

Table 3. Comparison of mortality for
all treatment groups

 

Treatment Groups % Mortality
15 second twilight 12.6%
““3 ”C“ 15 minute twilight 16.97.
15 second twilight 14.5%
Short cycle 15 minute twilight 14.7%
Constant Darkness 19.2%

Conditions Constant Light 21-5%

 

 

13

Table 4. Summary of data comparing mortality
rates of all six treatment groups.

 

 

 

 

 

 

Treatment # Flies # Flies Total

Groups Eclosed Aborted # Flies
Long cycle 15 sec. twilight 734 106 840
(25 hrs. 9 min.) 15 min. twilight 698 142 840
Short cycle 15 sec. twilight 718 122 840
(22 hrs. 52 min.) 15 min. twilight 716 124 840
Constant Light 565 155 720
Conditions Darkness 582 138 720

 

 

 

 

II. Effects of light cycles on eclosion. Differences in
eclosion rates were tested by comparing the number of flies eclosing
before and after 269 hours of incubation. Two hundred sixty-nine hours
is the best data point for dividing the bimodal eclosion curves of the
light cycle groups as can be seen in Figures 3 and 4. Two hundred
sixty-nine hours is also closest to the means of the constant light
and constant darkness treatment groups (Figure 5). For these reasons,
269 hours was chosen as the most representative point for dividing
all curves for :X? analysis of eclosion rates.

Three independent comparisons of eclosion rates were made of the
four light:dark treatment groups. Since each group was subdivided
into flies eclosing before and after 269 hours of incubation, a total
of seven independent comparisons could have been made. The other
comparisons are not relevant (e.g. those made by combining before and
after 269 hour groups) and furthermore do not yield significant
differences. The comparisons chosen were 1) long cycle versus short
cycle, 15 second twilight, 2) long cycle versus short cycle, 15 minute
twilight, and 3) fifteen second twilight versus 15 minute twilight. A

fourth independent comparison, constant light versus constant darkness

FLIES ECLOSING/4 HRS.

FLIES ECLOSING/4 HRS.

400

300

200

100 -

1111'

l

lllllll

 

400‘

I‘ll

D.)

O

O
l

N
c:
c:
1....

III

...:
O
O
l

Llll

 

.—..——.-_-_T

14

15 Second
Twilight Transition

 
  
 
  
 

.._.-—-short c cle
N=718 + 2

-— -—-long cycle
N=734 d'+ 5?

'“N/ \lgt

253257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

Eclosion curves of long and short cycle

/ \
I, ' .\\
l N \ . ”.\
I \ /\s/ “e~~
O . ‘

light:dark treatment groups, using a 15 second
twilight transition (males and females combined).

15 Minute
Twilight Transition

° —-o--short c cle

N-716 + 3
— o--‘long cycle

N=698 6+ 3

3..
I I ‘ '

 

253 257 261 265 $9 273 277 281 285 289 293
HOURS OF INCUBATION

Eclosion curves of long and short cycle

light:dark treatment groups, using a 15 minute
twilight transition (males and females combined).

15

 

: Consta t C d't’
32' 400 j n on 1 ions
m : -.u-..constant liift
-¢ * N=565 53+
:3 300‘:
a j .——.——.constant darkness
U) —1 N=582
a zoo -_+ 30’ 9
0 g o
m ‘1 I”:<\/’
(L3 100 A //./ \\o
H 1 / \\ '
'4 / o \\
m ‘ .
l . ’ll’o/ \\\:—>-s _¥
._1=:‘3—-——'" , ' T I i ;

 

I I
253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

Figure 5. Eclosion curves of constant

light and constant darkness treatment
groups (males and females combined).

was also tested. It has already been shown that the comparison, cyclic
conditions versus constant conditions, exhibits a significant differ-
ence in mortality. This comparison in terms of eclosion rate is not
valid because eclosion rate and mortality may not be independent; that
is, differences in eclosion may or may not be attributed to different-
ial mortality.

Of the four comparisons tested, all exhibit significant differ-
ences (P (0.01). These comparisons and their associated 12 values
are listed in Table 5. Table 6 is a summary of the data used for
comparing eclosion rates of the light:dark cycle treatment groups.

Flies incubated under short cycle light regimes eclosed signif-
icantly earlier than flies incubated under long cycle regimes
(P<0.0l). This is true for both 15 second and 15 minute twilight

transition groups.

16

Table 5. Summary of x2 eclosion rate statistics

 

 

Treatment P Relative
Groups x Value Rates*
-Long versus short, 15 second 2
twilight transition I (1)=159.648 P<.001 Short (Long
Long versus short, 15 minute
twilight transition 12(1)=273.817 P(.001 Short<Long
15 second versus 15 minute
twilight transition x2(1)=602.242 P(.001 15 min.<15 SEC.
Constant light versus 2 Const. light<.
constant darkness X. (l)= 45.935 P<-001 Const. darkness

 

 

 

 

*Short<Long = flies incubated under short cycles eclose before flies
incubated under long cycles.

Table 6. Summary of data used for comparing eclosion
rates of long and short cycle groups*

 

 

 

Sex Long Cycle Short Cycle Twilight
(25 hr. 9 min.) (22 hr. 52 min.) Transition
(269 hr1.j>269 hr. (269 hrT>269 hr.
68' 5 331 9 310
$3. 24 374 194 205 15 second
d'd‘ + 34 29 705 203 515
do“ 53 280 227 98
$4 217 148 358 33 15 minute
66' + $4 270 428 585 131

 

 

 

 

*(269=number of flies eclosing before 269 hours of incubation.
>269=number of flies eclosing after 269 hours of incubation.

Table 7. Summary of data comparing eclosion rates of
long and short cycle treatment groups*

 

 

Sex Constant Light Constant Darkness
.5252 fits.* [2262 hi§.* (262 hrs.* L>269 hr§.$
60" 112 149 63 232
1* 251 53 195 92
66 + $¥ 363 202 258 324

 

 

 

*(2698number of flies eclosing before 269 hours of incubation.
>269=number of flies eclosing after 269 hours of incubation

17

The bimodality so evident in eclosion curves of the light cycle
groups is not evident in the constant light or constant darkness
treatment groups. The eclosion curves of the constant light and
constant darkness treatment groups are significantly different
(P4C0.01, see Table 5). Flies incubated in constant light eclose
earlier than flies incubated in constant darkness. Table 7 is a
summary of the data used for comparing eclosion rates of the constant
condition treatment groups.

III. Effects of twilight transitions on eclosion. Flies incu-
bated with a 15 second twilight transition eclosed significantly
later than flies incubated with a 15 minute twilight transition
(P<0.01). Data of the long and short cycle treatment groups for a
given twilight treatment were combined to form a single twilight
transition group. In this way, biases imposed by differences in long
or short cycle treatments are accounted for.

IV. Effects of sex on eclosion. Males incubated under 15 second

 

twilight did not conform to the pattern exhibited by all other light:
dark treatment groups analyzed as separate sexes (see Figures 6-9).
That is, males incubated under short cycles (15 second twilight) did
not eclose earlier than males incubated under long cycles (15 second
twilight). In this case, there is very close agreement between the
long and short cycle eclosion curves (Figure 6). In all other treat-
ment groups, including females incubated under 15 second twilight,
both males and females did eclose more rapidly when incubated under

short cycles.

18

Bimodality is lacking in the male long cycle groups for both 15
second and 15 minute twilight transition groups. All other light
cycle treatment groups show bimodality.

These data cannot be analyzed for mortality of separate sexes

because the sex ratio of eggs was not known.

FLIES ECLOSING/ 4 HRS.

FLIES ECLOSING/ 4 HRS.

19

J

/.

.\ 15 Second
\ Twilight Transition

N

O

O
14

 
 
 
   

H
U1
0

1-1
0
O
-lllllllllllllllllll

_.— long cycle
N=336 a”

\——o--short cycle
\. N=3l9 or

 

OI \

1/43. \
I a
'0‘- \’__=—§_.‘_

 

o
I

253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

. ---.‘ '

 

Figure 6. Eclosion curves of long and short cycle
light:dark treatment groups, using a 15 second
twilight transition (males only).

   
  
 
 
  

15 Second
Twilight Transition

 

long cycle
N=398 9

--'--short cycle
N=399 g

 

1 .____,é:__,.i~‘u . ¥:>~4

253 257 261 265 269 273 277 281 285 289. 293
HOURS OF INCUBATION

 

Figure 7. Eclosion curves of long and short cycle
light:dark treatment groups, using a 15 second
twilight transition (females only).

FLIES ECLOSING/4 HRS.

FLIES ECLOSING/4 HRS.

20

 

 

 

1
2001 Constant Conditions

: ——u-—-constant darkness
150- N=295 0'

I . .——..__constant light
1004 ‘\\\\ N=26O 0'

I I/ ‘.\\ .
50d / . ‘0‘

‘ /

1 ,J/

‘—a---n~:— ref’ : . A*:-—=«==:~--1—

253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

Figure 10. Eclosion curves of constant light and
constant darkness treatment groups (males only).

 

200— Constant Conditions

1

: ._—..—.constant darkness
1501 N=287

: . ..-.-..constant light
100 j I, \\ , N=305

j // oX‘o\

_ 7’ \\ .

501 ll. \\\
: // .\\
l_‘r===£/’. \:‘~~3§ .
1 V I I I

 

 

I l T I
253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

Figure 11. Eclosion curves of constant light and
constant darkness treatment groups (females only).

FLIES ECLOSING/4 HRS.

FLIES ECLOSING/4 HRS.

21

 
  
 

 

 

  

 

 

 

200 5 .
3 15 Minute
‘ ° Twilight Transition
150 .2
: ‘——--——long cycle
: N=333 d”
100 — °
: I, \\\. ' __.._.. short cycle
- y \ N=325 O”
. / \\
50 ‘1 I, \.”’°\‘\.
3 I
. .==—=2~—-.—-./. . , . -- -
253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION
Figure 8. Eclosion curves of long and short cycle
light:dark treatment groups, using a 15 minute
twilight transition (males only).
200'5 15 Minute
- Twilight Transition
150 J -—-—long cycle
2 N=365
100-: \ __.—-short c cle
. \ N=39l
I \
.1 \
50 _ \
: .:xf/’ /' '
.. . \\ 1” \\\§’:.\

I I T I I

T I I v I
253 257 261 265 269 273 277 281 285 289 293
HOURS OF INCUBATION

Figure 9. Eclosion curves of long and short cycle
light:dark treatment troups, using a 15 minute
twilight transition (females only).

DISCUSSION

1. Effects of light cycles and twilight transitions on mortality.
Mortality was significantly higher for flies incubated in constant
conditions than for flies incubated in light:dark cycles (P‘<0.01).

No significant differences were noted between twilight transition
treatments or the long-short cycle treatments (P<0.01). Detrimental
effects of constant light have been recognized for some time. Pitten-
drigh (1960) tested the effects of constant light on the penetrance

of the recessive allele tug, responsible for melanotic pseudotumors

in Drosophila. The penetrance of tug dropped from 90% to 40% in the

 

first generation subjected to constant light. Differential mortality
of flies carrying the tug allele must be assumed.

In the same paper (1960, p. 170) Pittendrigh states, "There is
no direct evidence that constant darkness is detrimental to anything
but green plants." My data on mortality of fruit flies incubated in
constant darkness presents direct evidence for detrimental effects to
animals.*

These detrimental effects may result from a desynchronization of
*Another line of evidence is given by Riesen (1951). Riesen reports

that chimpanzees reared in darkness do not have normal visual capac-
ity. It is further shown that chimpanzees reared with their heads

enclosed in translucent domes exhibit the same abnormalities, imply-
ing that it is the lack of patterned vision and not darkness per se.

However, since constant darkness excludes patterned vision it must be
considered pathogenetic in its own right.

22

23

circadian organization. For example, peak times of sodium and potas-
sium excretion bear a fixed relationship when man adheres to a 24-hour
activity cycle. If the period of the activity cycle is changed,
sodium excretion follows the new cycle, but potassium excretion is
much more resistant to change and adheres to the intrinsic 24-hour
cycle. In this case the constituent rhythms of sodium and potassium
excretion have become desynchronized (Lobban 1960). Desynchrony can
be thought of as a change in the juxtaposition of physiological
events,inc1uding those important in development.

There is ample evidence for adult animals, as well as for plants,
that internal rhythms of sodium, potassium, and water excretion, body
temperatures, etc., can be desynchronized by l) constant conditions
(Todt 1962, Halberg and Barnum 1961), 2) reversed or shifted light
cycles (Lobban 1965), and 3) cycles greater or less than 24 hours
(Lobban 1960). Desynchrony as an explanation of the detrimental
effects of constant conditions may be tested using desynchronizing
stimulations other than constant conditions. For example, if random
light cycles or regular cycles that differ radically from 24 hours
are given, the constituent rhythms will become desynchronized (Lobban
1960). Assuming that desynchrony becomes greater in proportion to
the difference between exogenous synchronizers and the innate 24-hour
periodicity, one would predict increased mortality in related propor-

tions.

24

II. Effects of light cycles gn_ec1osion. Eclosion occured
earlier for flies incubated under short cycles than for flies incu-
bated under long cycles (PM(0.01). This supports the hypothesis that
developmental time is related to the length of exogenous oscillations.

Light cycles do phase developmental events. This is apparent in
the differences between eclosion curves of treatment groups subjected
to light cycles or maintained in constant conditions. Those flies
maintained in constant conditions have unimodal eclosion curves,
while flies incubated in cyclic conditions exhibit bimodal patterns
of eclosion. Bimodal patterns of insect eclosion and other circadian
functions, such as running-wheel activity of rodents and general
activity of birds, are common in the literature (Aschoff 1966).
Bimodality in eclosion curves may be attributed to the phenomena of
"gating" (Minis and Pittendrigh 1968). This phenomena implies that
the fly may eclose only at specific phase intervals of the endogenous
circadian cycle. These intervals are referred to as gates. If the
fly is not ready for the first gate he is obligated to wait for the
next one. It would appear that for the cyclic conditions of this
experiment, two eclosion gates exist within the final circadian
period, the centers of which are separated by approximately 12 hours
or one-half a circadian cycle (see Figures 3, 4, and 6 through 9).
The absence of bimodal eclosion curves in the constant condition
groups indicates that the circadian cycles (and therefore the gates)
of individual flies were not in synchrony as a population. The lack
of synchrony is to be expected in the absence of a common synchronizer

such as a light cycle.

25

Flies incubated in constant light eclosed significantly
earlier (P410.01) than flies incubated in constant darkness. In
absence of exogenous cycles, the endogenous cycle is free running.
Aschoff's Rule* would predict shorter free-running rhythms in con-
stant light. Therefore, one would predict earlier eclosion for
the constant light treatment group, based on the hypothesis stated
on page 1.

The constant conditions were included for comparisons between
eclosion of cyclic treatment groups; however, significantly higher
mortality rates of constant condition groups preclude this compari-
son. Mortality and eclosion rates may not be independent. If, for
instance, "weak” flies emerge later, and "weak" flies are differ-
entially selected against, then flies in treatment groups with higher
mortality would appear to eclose earlier even though the develop-
mental rate of any given fly had not been affected.

III. Effects 2f_twilight transitions on eclosion. Short cycle
groups eclosed before the long cycle groups in both the 15 second and
15 minute twilight groups. Both twilight transition groups exhibited
bimodal distribution and synchronization to the imposed light cycle.
A very real difference exists between these twilight treatment groups
however. This difference is best expressed as a delay of 4 hours in
the bimodal eclosion peaks and a lower first peak for both 15 second

twilight long and short cycle treatment groups when compared with the

*See note at bottom of page 1.

26

15 minute twilight treatment groups. This results in a significant
difference in the means of the two twilight groups (P (0.01). A
comparison of Figures 3 and 5 will also show that eclosion of the

15 second twilight group is also later than either constant darkness
or constant light. Statistical analysis of this comparison is ex-
cluded by the set of contrasts chosen; however, statistical compari-
son is hardly necessary.

It is clear that the length of twilight transition is the vari-
able involved, since both long and short cycle boxes are affected
similarly. Both sets of cyclic treatment boxes were subjectedto
identical environmental conditions of temperature, humidity, light
intensity, barometric pressure, and all other parameters except twi-
light transitions. Length of light cycles is also accounted for since
boxes controlled by both the long and short cycle timer produce 15
second twilight transitions.

The obvious question is: what aspects of twilight transitions
caused the shift to later eclosion in the 15 second twilight group
relative to the 15 minute treatment group? The answer is not
apparent, but three possibilities have arisen. They are:

1) A difference in total 11ght energy, Examination of
Figure 12 shows that more light energy reaches the flies at dawn
under the 15 second twilight transition, but this is exactly offset
by more light energy reaching the flies with the 15 minute twilight
at dusk. There is then, no difference in total light energy and

this possibility is unrealistic.

27

 

 

W///, :i/////W.m

Figure 12. Schematic representation of the twilight tran-
sitions used. Point A represents the onset of dawn. Point aS
represents the culmination of the 15 second dawn and am, the
culmination of the 15 minute dawn. Point B represents the

onset of dusk; bS represents the culmination of the 15 second

dusk and bm the culmination of the 15 minute dusk. The

shaded area represents the light phase and the cross-hatching

indicates the portions of the cycle that differ under the

different twilight transitions. The ratio between 15 second

and 15 minute twilight transitions is 1:60. Twilight transi-

tions illustrated in this figure are 1:10.

2) A difference in photoperiod. For any light intensity below
one-half "full light" the 15 minute twilight regime presents a longer
photoperiod and for any intensity above this point the 15 minute
regime presents a shorter photoperiod. Photoperiods for the light in-
tensity one-half way through a transition are equal for both twilight
groups. Differences in the photoperiod of the minimal effective light
intensity may be the factor involved, but this can only be determined
by further research using different photoperiods and equal twilights.

3) A difference in stimulus length. Figure 12 illustrates a
longer, more gradual transition for the 15 minute twilight group than
for the 15 second twilight group. This is the most obvious difference
between the two twilight treatments. The manner in which this might

account for the difference noted between twilight treatments is not

obvious.

28

Kavanau (1962) has shown that the period of activity, feeding
and drinking cycles of Peromyscus can be manipulated more radically
with twilight transitions than with the traditionally used 0N-0FF
light transitions. Perhaps a similar principle is responsible for
both observations. Kavanau does not discuss relative stimulus values
of twilight versus an ON-OFF light transition,except to indicate that
twilight simulates field conditions.

The effects of twilight on the entrainment of endogenous cycles
is probably the most significant parameter of light and certainly one
of the least studied or understood.

IV. Effects g£_§gxugn eclosion. Figures 6 through 11 Show
earlier eclosion of females as compared with males. The more rapid
development of females is well documented and supported by Bakker and
Nelissen (1963), Bonnier (1926), Poulson (1934), and Powsner (1935).
In Bakker and Nelissen's study (1963), the sex difference was not
evident in the egg and larval periods, but was confined to the pupal
period. They also noted that a strong correlation existed between
weight and time of eclosion. Heavier males eclosed before lighter

males and heavier females before lighter females. Female Drosophila

 

weigh.more than males. It appears that weight may be an important
factor in determining the earlier eclosion of females. Bakker and
Nelissen report that peak female eclosion averages 7.5 hours earlier
than male eclosion.

With the exception of the data collected under 15 second

29

twilight regimes, analysis on the basis of sex does little to alter
the relationship between long and short cycle groups. With the 15
second twilight transition regimes, it is apparent that the bimodal
nature of the eclosion curves has changed radically and that the

females are responsible for any contribution toward bimodality.

OVERVIEW

Prior to this thesis, a functional role of exogenous circadian
cycles in ontogeny had not been described. Data presented here
support the theory that exogenous circadian cycles act as a chron-
ometer for development. That is, the more rapidly the cycles occur
the more rapidly development proceeds. Realistically, limits are
to be expected on the extent to which development can be altered.
These data are not adequate for speculating on these limits.

In addition, it appears that the presence of such a chronometer
maximizes developmental success. In the absence of exogenous cycles,
mortality increases. It is possible that the absence of a chronometer
(i.e. light cycles) allows component rhythms of the developing organ-
ism to become desynchronized, and that such desynchrony is translated
into a derangement of developmental events. If the integrity of the
developmental sequence is lost, abnormal deveIOpment and increased
mortality can be expected.

Both constant darkness and constant light have been shown to
cause desynchrony of constituent endogenous rhythms (Todt 1962).
Constant light is the stronger desynchronizer and has been shown by
others (Pittendrigh 1960) to be pathological. The significant differ-
ence in mortality rates, found between flies incubated in constant
conditions or in light cycles, is a demonstration of the pathological

effects of constant darkness as well as constant light.

30

31

Light cycles longer or shorter than 24 hours have also been
shown to desynchronize constituent endogenous rhythms (Lobban 1960).
In this case, one would expect mortality of flies incubated in the
long and short cycles described here to be greater than mortality of
flies incubated in 24-hour 1ight:dark cycles. This question can not
be answered from these data because a 24-hour 1ight:dark cycle was
not run concurrently.

The two major concepts developed in this thesis are: 1) exo-
genous circadian cycles act as a chronometer for develOpment, and
2) exogenous circadian cycles, by synchronizing development, maxi-
mize developmental success. The hypothesis* I set out to test is
the basis of the first concept. The second concept became evident
from the mortality data. The concept that exogenous cycles maximize
developmental success encompasses the idea that light cycles act as
a chronometer and is the more interesting of the two. Indeed, it
implies that the presence of exogenous circadian oscillations are
important to survival.

These concepts generate several related questions. The series
of questions I am most interested in are:

1) What conditions produce the greatest difference in mortality?

2) At what stage of development does highest mortality occur
under desynchronizing conditions?

3) What abnormalities lead to abortion?

*The developmental period, from oviposition to emergence of the
Drosophila melanogaster imago, is longer when flies are incubated
in long circadian 1ight:dark cycles, than when they are incubated
in shorter circadian light:dark cycles.

32
4) What are the processes controlling normal or abnormal
development?
5) How are these processes tied into exogenous circadian

cycles; and, how do they differ under synchronizing and
desynchronizing conditions?

The exceptional case of males subjected to 15 second twilight
transitions (Figure 6) suggests sex and twilight as variables requir-
ing more attention than can be given from these data. A considera-
tion of twilight and sex in conjunction with the theory of "gating"
(see p. 24) offers a tentative explanation of the late eclosion of
males subjected to a 15 second twilight. Two possible gates are
indicated by the other curves of flies subjected to cyclic conditions.
Slower eclosion rates are correlated with both sex and 15 second twi-
light. It is conceivable, therefore, that only a few males were
competent to eclose at the first gate, whereas practically all were
competent by the second gate. This explanation is unsatisfactory,

but is the best available at this time.

SUMMARY

The two major concepts developed in this thesis are: l) exogen-
ous circadian cycles act as a chronometer for development, and 2) exo-
genous circadian cycles, by synchronizing develOpment, maximize
developmental success.

The first concept is based on more rapid development of Drosophila,
during the period oviposition to eclosion, when incubated in shorter
light:dark cycles.

The second concept is based on lower mortality rates for flies
incubated in light cycles as compared to constant light or constant
darkness. It is probable that increased mortality results from
desynchrony of constituent rhythms in constant conditions.

Males eclose later than females and flies incubated under 15
second twilight transitions eclose later than flies incubated under
15 minute twilight. These statements may be made with certainty;
however, further research is required before concepts of functional

roles in ontogeny can be developed for these parameters.

33

34

Appendix I. Raw Data

The raw data is given in Tables 8 through 13. Each column
represents the flies eclosing in a given vial. The columns are
further divided on the basis of sex. Flies were collected at
4 hour intervals. The hours of incubation preceding each collec-
tion are indicated in the left hand column. The total number of
flies eclosing in any one time period is totaled across columns
and entered in the right hand column. The number of flies eclos-
ing in any one vial is totaled across time periods and entered
in the bottom row. The total of either sex eclosing in any one

vial is given in the next to the bottom row.

35
Raw data for the long cycle, 15 second twilight treatment

Table 8.

Shutterbox #1.

group.

 

Totals

79922

a

14/4*

734 .

 

 

7

0130063210000

0000007030040
121

100

 

0220008010000
22

000013M900110

102

 

1000057100000

 

 

 

 

 

 

 

 

0+ 6
A.
4f A.
5
0+ a.
5
Jo 9
A.
0+ .4
.5 1.1. ,omw
Jononunvnv1.9.s31.9.nv1.A.nvAu1i
as]. .5
r
e O+AUnu1lnunv1.926.iAU1lnvnvo,
bl... 1A. 69
m .1
u 10AUAUnununuqsth.1.nv9.9.nonU11
N 1.2 5
1
a 9.21100692200003
mu.3 9.9. ,omw
4d.111nununus30993nv1.nv1.n.131.
9. A.
0+o215nu9.nv9.131.nvnvnvnvn.a.
9. 9.9. Rem“
xv.UAUAUAUnuansnuqsnunut.nunu
o. .4
0+nu1ioznunvqenusenvnvnv1.n.nu
1.. 1../... 6%
-.f 1.nvnvnunvA.7.nv1.1.nvnvnu.411
9.1. .4
n
o
,t.1
0t 1
a 3715937159397X8
s.D .5.5,o,o,01/vlnununuoanu1. e.1
r u 9.9.9.9.9.9.9.9.9.9.9.1313 8 v
a... 7/
HI 22

 

 

 

 

 

Raw data for the long cycle, 15 minute twilight treatment

Table 9.

Shutterbox #2.

group.

 

Totals

20
92
69
288
112
3/4*
1/2*

698

 

Vial Number

10

14

2661133000100

1001441400000
21...

88

 

13

1620093010000
1 11..

nv1.1.nuRuo,RVI.nvnvnvnvnv
9.1.

110

 

12

2151604000000
11 2

1110353201000
21

106

 

11

1511573010000
11 l

0000279101010
21

105

 

7971771100000
1..

0212168101010
1

83

 

Mu7.nunvnvnununu

1111600300001
21

24302
11

103

 

 

o’go’io’go’gozgo’ifi

 

2420423010000
11 1

0201089400010
12

55 48 44 59 33 50 51 54 47 59 58 52 45 43

103

 

 

Hours of

Incubation

 

253
257
261
265
269
273
277
281
285
289
293
309
317

 

E/sex

 

ZZ/vial

 

 

 

time periods (i.e. these periods are greater

* # of flies/# of 4 hr.

than 4 hrs.)

36
Table 10. Raw data for the short cycle, 15 second twilight treatment

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

group. Shutterbox #3.
Hours of Vial Number
Incubation 15 16 17 18 19 20 21 Totals
61.9013 a»? 6152 o’fii M3 67$

253 o o o o o o o 1 o 1 o 1 o o 3
257 o o 1 1 1 3 o 3 o o o o 1 1 11
261 o 9 o 16 1 18 o 8 o 7 o 6 o 5 70
265 2 21 o 9 1 12 o 13 o 10 o 11 o 11 90
269 1 2 o 1 o 3 o 5 o 5 1 8 o 3 29
273 3 9 3 5 3 3 4 12 3 10 8 15 2 11 91
277 32 15 3o 24 29 18 32 19 42 9 36 11 22 29 348
281 9 o 8 1 1 1 7 2 4 2 6 1 1o 5 59
285 2 o 2 1 1 o o o o 3 1 o 2 o 12
289 o o 1 o o o o o o o o o o o 1
293 o o o o o o o o o o o o 1 o 1
309 1 o o o 1 1 o o o o o o o 0 3/4*
317 o 0 o o o o o o o o o o o 0 0/2*

Sf7sex 50 56 45 58 39 6o 43 63 52 44 52 53._38 65 718

A 2/v1a1 106 103 99 106 96 105 103

 

Table 11. Raw data for the short cycle, 15 minute twilight treatment

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

group. Shutterbox #4.
Hours of Vial Number
Incubation 22 23 24 25 26 27 28 Totals
M$M$J$J$a$ofl$w$

253 1 0 l 1 2 6 O 0 1 l 1 4 O O 18
257 2 26 0 12 9 30 1 25 O 14 0 26 1 27 173
261 18 24 13 25 24 13 13 26 9 29 19 22 3 16 254
265 9 5 15 4 9 1 12 4 15 5 14 4 4 7 108
269 3 0 9 0 1 0 4 0 6 O 3 1 5 0 32
273 9 1 1 13 3 0 9 1 11 4 3 0 11 3 69
277 3 0 4 0 4 1 6 2 7 2 3 1 5 0 38
281 4 0 O 0 0 0 2 1 0 0 3 0 2 O 12
285 O O 1 O O 1 O 0 1 0 0 0 1 0 4
259 0 0 0 0 O 0 0 0 1 0 0 0 0 0 1
293 0 0 0 O 0 0 0 0 0 0 0 0 0 0 0
309 l 0 0 2 1 0 0 1 0 0 O 0 1 0 6/4*
317 0 0 0 0 1 0 0 0 O 0 O O 0 0 1/2*

X/sex 5056 44 57 54 52 4760 5155 46 58 33 53 716

Z/vial 106 101 106 107 106 104 86

* # of flies/# of 4 hr. time periods (i.e. these periods are greater

than 4 h

rs.)

 

 

37

Table 12. Raw data for constant dark treatment group.

Shutterbox #5.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hours of Vial Number
Incubation 29 3O 31 32 33 34 Totals
0" f- 6» 4 «v 41 0" 41 o" 19 62' +9
253 l 2 O l 0 l 0 0 l 2 l l 10
257 1 0 l 0 O 0 0 0 O 0 0 1 3
261 O 3 0 1 O O O 2 O 5 1 4 16
265 0 7 3 9 2 13 1 ll 0 l9 1 16 82
269 6 22 4 14 8 8 15 22 14 13 3 18 147
273 9 18 20 15 14 6 24 8 26 5 17 9 171
277 22 5 13 2 13 2 10 2 11 1 10 5 96
281 5 4 2 1 l 3 2 3 6 0 4 1 32
285 2 O 0 1 2 O 1 O 1 O 2 1 10
289 2 O l O 0 0 O O l O O 0 4
293 2 0 O O 1 O 1 O 0 O O 0 4
309 0 O 2 O O 0 O 0 1 O 2 O 5/4*
.2? 317 0 0 1 0 0 0 0 0 0 0 1 0 2/2*
T/sex 50 61 44 44 41 33 54 48 61 45 42 6_. 582
_2;/v1a1 111 91 74 102 106 98
Table 13. Raw data for constant light treatment group.
Shutterbox #6.
Hours of
Incubation 35 36 37 38 39 40 Totals
6" +°~ cf' 2 o" 3 a? 3 o” 3 6" 3
253 1 2 O 2 O l O l O 0 2 1 10
257 l 3 0 0 O 0 0 3 l 1 0 3 12
261 0 4 0 9 O 13 0 10 0 13 0 5 54
265 4 13 3 20 2 15 2 25 4 13 7 21 129
269 3 16 17 ll 19 4 20 11 16 22 10 9 158
273 19 7 15 4 8 6 ll 9 15 3 9 3 109
277 13 4 7 3 3 1 10 2 9 0 11 2 65
281 0 0 5 l 1 O 2 1 2 O 2 l 15
285 1 4 0 1 1 0 0 0 0 0 2 0 9
289 0 0 l O 0 0 O O 0 0 0 0 1
293 0 0 O 0 O 0 O O l O O 0 l
309 0 0 0 0 O 1 0 0 0 0 0 0 1/4*
317 O 0 O O 0 0 O O 1 0 0 0 1/2*
§;/sex 42 53 48 51 34 41 45 52 49 52 43 45 565
2‘, vial 95 99 75 107 101 88

 

 

 

 

 

 

 

 

 

 

* # of flies/# of 4 hr. time periods (i.e. these periods are greater

than 4 hrs.)

 

 

38

Appendix II. Density—dependent developmental rates.

Pilot work for the experiments described in this thesis was
done without controlling density of flies developing in each vial.
A number of adult flies oviposited in each vial for a four hour
period and were removed after that time. The number of eggs ovi-
posited in each vial varied widely as indicated in Figures 13 and
14.

It is clear from these figures that developmental rates are
density dependent; and that higher densities result in retarded

eclosion.

39

Figure 13. Density dependence of eclosion rate for
flies incubated in short light:dark cycles (22 hr.

52 min.). Each data point represents one vial.

The number of flies developing from egg to imago

in that vial is represented on the absisa, and the

percentage of the total number eclosing by 249 U3.

252 hours of incubation is indicated on the ordinant.

Figure 14. Density dependence of eclosion rate for
flies incubated in long light:dark cycles (25 hr.

9 min.). Other conditions and information same as
given for Figure 13.

100

90

80

% ECLOSED
u: ¢~ 1m ox \1
<3 c: c: c> c:

N
O

10

100

90

80

Z ECLOSED
1n ox ~u
c: <3 c:

b
O

30

20

10

40

 

 

 

 

PO. . .... . . o .
' ‘ 1 . . Figure 13
46 L 86 ' 120 160 200 240 J7 280
NUMBER OF FLIES/VIAL
P . . '0‘ .. .: 0.. .
F 0 o .0 .
: . 'o Figure 14
A 40 80 120 160 200 240 280

NUMBER OFFFLIES/VIAL

BIBLIOGRAPHY
Aschoff, J., 1958. Tierische Periodik unter dem Einfluss von Zeit-
genbern. Z. Z. tierpsychol. 15(1):l-30.

, 1966. Circadian activity patterns with two peaks.
Ecology, 47:657-662.

 

Bakker, K. and F. X. Nelissen, 1963. On the relations between the
duration of the larval and pupal period, weight and diurnal
rhythm in emergence in Drosophila melanogaster. Ent. Exp. &
Appl. 6:37-52.

Bonnier, G., 1926. Temperature and time of development of the two
sexes in Drosophila melanogaster. J. Exp. Biol. 4:186-195.

Brett, W. J., 1954. Persistent diurnal rhythmicity in Drosophila
emergence. J. of Tenn. Acad. of Sci. 29:176.

Demerec, M. (ed.), 1950. Biology of Drosophila. John Wiley & Sons;
New York, N. Y.

Halberg, F. and C. P. Barnum, 1961. Continuous light or darkness
and circadian periodic mitosis and metabolism in C and D8 mice.
Am. J. Physiol. 201 (2):227-230.

Harker, J. E., 1965a. The effect of a biological clock on the devel-
Opmental rate of Drosophila pupae. J. Exp. Biol. 42:323-337.

 

, 1965b. The effect of photoperiod on the developmental
rate of Drosthila pupae. J. Exp. Biol. 43:411-423.

Hellbrugge, T., J. E. Lange, J. Rutenfranz and K. Stehr, 1964. Circa-
dian periodicity of physiological functions in different stages
of infancy and childhood. Ann. of N. Y. Acad. of Sci., 117:361-
373.

Johnson, L. G., 1966 Diurnal patterns of metabolic variation in
chick embryos. Biol. Bull. 131:308-322.

Johnson, M., 1939. Effect of continuous light on periodic spontan-
eous activity of white footed mice. (Peromyscus) J. Exper.
2001., 82:315-328.

Kavanau, J. L., 1962. Activity patterns on regimes employing artifi-
cial twilight transitions. Experimentia 18(8):382-384.

41

42

Lobban, M. C., 1960. The entrainment of circadian rhythms in man.
Cold Sp. Harb. Symp. on Quant. Biol., 75:325-332.

, 1965. Dissociation in human rhythmic functions. In
Aschoff, J. (ed.); Circadian Clocks. Amsterdam: Horth-Holland
Publ. Co., 1965. 219-227.

Minis, D. H. and C. S. Pittendrigh, 1968. Circadian oscillation

controlling hatching: its ontogeny during embryogenesis of
a moth. Sci. 159:534-536.

Petren, T. and A. Sollberger, 1953. Die 24 stunden Rhythmik des
Leberglykogsres bei Huhnerembryonen und Kuken verschiedenen
Alters nebst Studien uber die Unabhangrigkeit der Rhythmik
von ausseren Faktoren. Act. Med. 145:54-66.

Pittendrigh, C. L., 1960. Circadian rhythms and the circadian organ-
ization of living systems. Cold Sp. Harb. Symp. on Quant.
Biol., 25:159-181.

Poulson, D. F., 1934. Times of development of the two races of
Drosophila pseudoobscuro. J. Exp. Zool. 68:237-245.

Powsner, L., 1935. The effects of temperature on the durations of

the developmental stages of Drosophila melanogaster. Physiol.
2001., 8:474-520.

Riesen, A. N., 1951. Chicago Med. Sch. Quart., 13:17-24. Cited
from Hebb, D. 0., 1958. A Textbook of Psychology, p. 130.
W. B. Saunders Co., Philadelphia & London.

Rutenfranz, J., 1961. The development of circadian system functions
during infancy and childhood; in S. J. Fomon (ed.) Circadian
Systems. Report of the thirty-ninth Ross conference on pedia-
tric research., p. 38. Columbus: Ross Laboratories.

Shutze, J. V., J. K. Lauber, M. Kato and W. O. Wilson, 1962. Influ-
ence of incandescent and coloured light on chicken embryos
during incubation. Nature 196:594-595.

Thomas, J. B. and D. J. Pizzarello, 1967. Blindness, Biologic
rhythms, and menarche. Obstetrics-Gynecology, 30:507-509.

Todt, D., 1962. Untersuchungen uber Offnung und Anthocyangehalts-
veranderunger dur Bluten von Cichorium intybus im Licht-Dunkel-
Wechsel und unter Konstanten-Bedergunger. Z. Bot. 1-21.

MICHIGAN STRTE u“1v
IIIII III III IIIIIIIIIIIII III III IIIIIIIIIII I2IIII5;
312931012

 

IRIIII

——~——<___.._