‘——— —. TEE 0; QQENY 0F COORDINATE!) LIME MOVEflENTS 1N PKLBL‘FEHED C'EECKSI EFFECTS. OF CENTRAL NERVQUS SYSTEM EAMFEILATIGNS ON TEE FEEQEEHCY AND PATTERNS OF MOYEHEHTS 4 That: for {in Dom”! 13h. D. MICBIGM 3TH}: mmm 7 » AEEeoM. Nikmdiwe ' 197.2 WMllllllfllflllllllmllfifll l/ LIBRARY 3 1293 01095 5031 Michigan State University This is to certify that the ‘_ ‘. . thesis entitled . N THE ON'I‘OGENY OF COORDINATE LIME MOVEMENI'S IN PRE-HATCHED CHICKS: EFFECTS OF CENTRAL NERVOUS SYSTEM MANIPULATIONS ON THE FREQUENCY AND PATIERNS OF mvmms. presented by Alfeo M. Nikund fire has been accepted towards fulfillment of the requirements for __Bh...n._degree in W ABSTRACT THE ONTOGENY 0F COORDINATED LINE MOVEMENTS IN PRE-HATCHED CHICKS: EFFECTS OF CENTRAL NERVOUS SYSTEM MANIPULATIONS ON THE FREQUENCY AND PATTERNS OF MOVEMENTS By Alfeo M. Nikundiwe This investigation is a two-part study designed to examine the de- velopment and control of coordinated limb movements in pre-hatched chicks, between day 19 of incubation and hatching. In the first part of the study, two developnental variables were examined: the temporal pattern of act- ivity and the magnitude of limb coordination.gIt has been shown by other workers that the alternation of legs (stepping), among other integrated movanents, does not occur at all until shortly after day 17. This stuck, therefore, attempted to ascertain whether some features in the develop- ment of leg coordination could still be shown during the last two days of incubation. The second part of the study dealt with various manipulations of the central nervous system, the purpose of vhich was to specify those (CNS) areas which funetion to subserve the alternate pattern of leg move- ments. To this end, different types of lesions were performed on differ- ent groups of embryos. The mid-thoracic transverse section (MT'I'S) aimed at excluding the possible influence of the brain from reaching and act- ing on the lumbosacral cord. The lumbosacral mid-saggital section (ISMSS) served to isolate that segnent of the spinal cord into left and right halves. This type of lesion not only was designed to uncover subtle Alfeo M. Nikundiwe reciprocal influence between each half of the cord, but it also enabled the experimenter to determine whether or not limb coordination could oc- cur in the absence of cross fibers. To answer the question as to whether peripheral feedback is involved in the alternate pattern of coordination, deafferentation of a group of embryos was accomplished by extirpation of the dorsal half of the lumbosacral cord (ISDF). In addition to these "basic" lesions, mid-saggital sections and deafferentation were carried out in conjunction with mid-thoracic transections on two other groups of embryos in order to further determine the brain/lumbosacral cord inter- action. These "double" lesions were termed mid-thoracic transverse sec- tion plus lumbosacral mid-saggital section (M’l'l‘S + ISPSS) and mid-thor- acic transverse section plus lumbosacral deafferentation (M'I'I'S + LSDF) respectively. All lesions were performed directly on the neural tube then the enbryos were 2&3 days old (stages 15-18, Hamburger and Hamilton, 1951). The eggs were left to incubate until day 19, at mich time each embryo had part of its shell removed from the blunt end so as to expose the legs. No regeneration was observed in any of the lesions. Leg movements were recorded by use of a polygraph recorder coupled with force-displacement transducers. Results demonstrated that neither the degree of limb coordination nor the absolute rate of movements manifested clear-cut changes during the two-day period. One conclusion drawn from these findings is that a high degree of limb coordination is alreacv well established by day 19. Animals with mid-thoracic spinal gaps were able to perform the al- ternate movements which were indistinguishable from those of the control group. However, activity level of the spinal animals was significantly Alfeo M. Nikundiwe lower than that of the controls. It was nonetheless concluded that the lumbosacral cord possesses the specific apparatus that is competent to bring about consistent alternation of legs. Embryos with mid-saggital sections exhibited a higher rate of activity than did the controls, but the pattern of movements was of the type in which the two legs moved simultaneously. From this experiment two conclusions were made: there exists reciprocal inhibition between the two halves of the cord segment; and that cross fibers are necessary in the execution of the alternate movements. Similarly, deafferentation caused an increase in the rate of coordinated movements, mile the resulting pattern of movements was of the simultaneous type. The results suggest that sensory neurons not only act to attenuate motor output, but they are also necessary for consist- ent alternation of leg movements. Dnbryos with the double lesions perform the simultaneous coordin- ated movements at much higher rates than the control groups. Since act- ivity level in the double-lesioned animal is also greater than that of the corresponding single lesion (i.e. M'l'l‘S + Isms > ISMSS; M'I'I'S + LSDF) LSDF), it was concluded that the brain inhibits neuronal activity occur- ring in the lumbosacral cord, but that this inhibition is best detected when the lumbosacral cord is disrupted (saggital section or deafferenta- tion). The results also further suggest that the brain and lumbosacral influences are additive. An alternative explanation which accounts for the tremendous increase in activity in the double-lesioned groups is to postulate that the brain does not directly inhibit the motor neurons in- volved in leg moments, but rather inhibits some other cells in the lumbosacral cord, those action, in turn, is to inhibit leg motor neurons. In essence, such a mechanism would be inhibition of inhibition. Were Alfeo M. Nikundiwe such a mechanism to operate, it would also account for the observed low- ering of activity in embryos with only mid-thoracic transections. P0881- ble mechanisms subserving the simultaneous pattern of coordination are discussed. THE ONTOGENY OF COORDINATED LINE MOVEMENTS IN PRE-HATCHED CHICKS: EFFECTS OF CENTRAL NERVOUS SYSTEM MANIPULATIONS ON THE FREQUENCY AND PATTERNS OF I‘DVEMENTS By Alfeo M. Nikundiwe A THESIS Submitted to Michigan State University in partial mlfillment of the requirements for the degree of IDCTOR OF PHILOSOPHY Department of Zoology 1 972 0 TABLE or CONTENTS HST OF TABLES 0..COOOOOOOOO0.0.0.0....OCOOOOOOOOOOOOOOOOOOOOOOOO I‘IST OF FIGms O0.00QCOOOCOOOCOOOIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO INTRODUCHON ANDREvIEWOF IJITERATUE .OOOOOOOOOOOOOOOO0.0.0.0... AppmaChes to the Study eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee General findings eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee QueStions flaked eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee DCperimen‘bal defign eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee MATERIALS AND IETHODS .0...OO...000......OOOOOOOOOOOOOOOOOOOOO... Premration Of efflbryos OOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO surgical Operations O...OOOOOOOOOOCOOOOOOOOOO0.00.0.0000... Recomj-ng OOOOOOOOOOCOOO0.0000......OOOOOOOOOOOOOOOOOOOOOOO measumments 0.0.0.0....00......00......OOOOOOOOOICOOOOCOOO RESUL'IS OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOO....00... Tanpoml pattem Of aetifity eeeeeeeeeeeeeeeeeeeeeeeeeeeeee Tmporal pattern or 1131b coordination eeeeeeeeeeeeeeeeeeeee Activity peak as a basis for analysis ..................... ACtiVity peak Cums eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee DISCIJSSION OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOIOOO0.00......00.0.00... Temporal pattern of coordinated movements ................. Effects of MTI‘S on the frequency and pattern of coordin- ated momts 0.0...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0... Effects of ISMSS on the frequency and pattern of coordin- ated moments OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Effects of ms + LSMSS on the frequency and pattern of coorflnated momenta O...OOOOOOOOOOOOOOOOOOOO0.0.00.0.0... Effects of deafferentation of the lumbosacral region on the frequency and pattern of coordinated movements ...‘..... SUMMARY OF AND CONCLUSIONS FROM LESION EXPERIMENTS .............. Frequency of coordinated movements ........................ Pattern of coordinated movements .......................... ii —I Q'QM‘ 1o 11 1h 15 16 1 7 1 7 1 9 21 30 30 32 3h 35 39 Mi 145 TABLE OF CONTENTS-- continued Page GWLW .00...00......COO...OOOOOOOOOOOOOOOOOOOOOOO0.0... 1‘79 BIBIIIOGMPIII 0.00000000000000000000000000COOOOOOOOOOOOOOOO0.00... 52 APPENDIX (Raw dam) 0..O....00..0.0.0.0....OOOOOOOOOOOOOOOOCOO... 58 iii Table 10. 11. 12. 13. 1h. 15. 16. 17. 18. LIST OF TABLES Experimental design ....................................... Summary of results ........................................ Raw frequency data for the control group .................. Raw frequency data for the M'l'l‘S group .... Raw frequency data for the ISMSS group .................... Raw frequency data for the LSDF group ..................... Raw frequency data for the MTTS + LSMSS group ............. Raw per cent data for the control group ................... Raw per cent data for the M'l'l‘S group ...................... Raw per cent data for the ISMSS group ..................... Raw per cent data for the ISDF group ...................... Raw per cent data for the MTTS + 13163 group .............. Raw activity peak curve data (frequency) for the control gmup 0.00.00.00.0000......0.00.00...OOOOOOOOCOOOOOOOOOO0.0 Raw activity peak curve data (frequency) for the M'l'l‘S group 00.0.0.0...00......OOOOOOOOOOOCOOO...OOOOOOOOOOOOOOOO Raw activity peak curve data (frequency) for the ISMSS gmup OOOOOOOOOOOOOO00.0.0.0...OOOOOOOOOOOOOOOOOOOOOOOOOOOO Raw activity peak curve data (frequency) for the LSDF gmup OOOOOOOOOOOO00.00.00...0.00.00.00.0000000000000000... Raw activity peak curve data (frequency) for the MTTS + ISMSS gmup OOOOOOOOOOOOOOOOOO...OIOOOOOOOOOOOOOOOOOOOO0.0. Raw activity peak curve data (frequency) for the MTTS + ISDF. group 0.0.0.000...OOOOOOOOOOOOOOOOOOOOOOIOCOOOOOOOOOO. iv 29 6O 61 62 63 6h 66 67 69 70 72 73 71: 75 76 77 CD LIST OF FIGURES hang A. B. 1. 2. 3. 10. 11. 12. C. SChema Of legions O00.....0...OOOOOOOOOOOOOOOOOOOOOOO0.00.. Block diagram of the microsurgical apparatus .............. Frequency of leg activity and per cent coordinated move- ments for the control group ............................... Frequency of leg activity and per cent coordinated move- ments for the mid-thoracic transverse section (M'l'l‘S) group. Frequency of leg activity and per cent coordinated move- ments for the lumbosacral mid-saggital section (ISMSS) gmup OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Frequency of leg activity and per cent coordinated move- ments for the lumbosacral deafferentation (LSDF) group .... Frequency of leg activity and per cent coordinated move- ments for the double 1881011 (MTTS + LSFBS) group eeeeeeeeee Control vs M'I'I‘S at activity peak .......................... Control vs LSMSS at activity peak ......................... Control vs LSDF at activity peak .......................... Control vs MTTS 4- ISTBS at activity peak .................. Control vs MTI‘S + ISDF at activity peak ................... LSMSS vs M'l'l‘S + LSMSS at activity peak .................... ISDF vs M'l'l‘S + ISDF at activity peak ...................... Block diagram of the proposed inhibitory mechanism .. ...... 17 17 18 18 1 9 21 22 23 21: 2S 2? 28 37 INTRODUCTION AND REVIEW OF LITERATURE A number of approaches have been employed in the study of patterns of coordinated movements. The type of strategy each investigator selects to use is determined by, among other things, the tools and expertise available to him, the nature of the problem under investigation, and of- ten by the type of specimen with which he works. The array of tactics used and the questions raised all share a common goal; that of attempt- ing to elucidate on the structuro-functional organization of the organism or of its parts. At the present time, the different approaches can be classified in- to four broad categories: the electrophysiological method, focal electric- al stimulation, gross lesion, and limb grafting and cord transplantation. Despite the superficial distinction among them, these tools overlap quite widely in application. The relative merits and demerits of each approach are all too obvious to the investigator. Lesion studies, for instance, are hardly the tools with which to study the integrative properties of the nervous system, but neither can the "unit" approach be said to tell much about the behavior of neurons at the population level. Nevertheless, discrete use of any one of these methods or a combination thereof, has enabled investigators to compile invaluable information about systems and subsystems and how they interact to produce whole behaviors . The insect neurophysiologist records the discharge patterns of leg motor neurons during normal walking in unrestrained animals (Ewing and Manning, 1966; Usherwood and Runion, 1970). The results of these investi- gations define fairly accurately, the normal patterns of motor neuronal activities which, in turn, must be described in terms of the properties and connections of the cells within the central nervous system and other neural activity in the various reflex pathways. Motor neuronal activity can also be investigated in dissected preparations by selective stimula- tion of the various constellations of peripheral receptors (Usherwood, Runion, and Campbell, 1968). Furthermore, it is possible to record spon- taneously generated activity of the motor neurons before and after deaf- ferentation (Pearson and Iles, 1970). By comparing these records and those obtained from freely walking animals, conclusions can be drawn about the existence of central programs and how they are modified by peripheral feedback. Recently, the technique of focal electrical stimulation of the brain with chronically implanted electrodes has been used to study simple inte— grative functions of the nervous system (Fraser-Howell, 1963 a, b; von Holst and Saint Paul, 1962, 1963; Vowles, 1961 ; Huber, 1965). The behav- ioral patterns which have been investigated mostly are locomotion, brea- thing, stradulation and flight (Wilson and Gettrup, 1963; Huber, 1962). These kinds of studies were not so much concerned with finding which brain centers or regions control what behavioral activities, as they were concerned with delineating the role played by the brain and other areas in influencing those areas directly involved with the execution of move- ment patterns themselves. The lesion method, by which a portion of neural tissue is surgic- ally removed or the continuity of the nervous system disrupted, has serv- ed to complement both electrophysiological and focal electrical stimula- tion studies and vice versa (Bethe, 1930; Reader, 1963; Hughes, 1957; Hamburger and Narayanan, 1969; Ten Gate, 1961; to mention only a few). 3 Daring the last decade or so, a sophisticated method of lesioning has been developed in which small areas of neural tissue are destroyed by use of electrocoagulation (Ballintijn, 1961; Vowles, 1958). Included in the lesion category is selective irradiation of neural tissue with gamna rays (Oppenheim et al., 1970). The majority of these studies dealt with vari- ous attempts to reconstruct the structural basis for control of behavior from what is alreacw known about the properties of the nervous system. One of the primary objectives as well as consequences of this approach has been to give the investigator an insight into, and a first approxima- tion of, the system with which he is working and on which to build fur- ther refinements. The pursuit to study development of motor patterns has led investi- gators to use embryonic preparations exclusively. The method which makes it possible to manipulate the embryo has been cord grafting and limb transplantation (Coghill, 1929; Carmichael, 1926, 1927; Detwller, 1936; Weiss, 19141 a, b; Szekely, 1963; and Hamburger, 1972). The results ob- tained from the experiments carried out by these workers suggest that in vertebrates, at least, the spinal cord segment at the limb level alone possesses the specific apparatus which enables the limb to move in a giv- en manner. Furthermore, Szekely (1967) and Hamburger (1972) were able to demonstrate that the type of spinal cord segment, thether brachial, thor- acic, or lmbosacral, determines the type of limb coordination to be ex- hibited. Thus, in a 19-day chick unbryo, only the brachial segnent 1:111 permit simultaneous flapping of the wings, while the lumbosacral cement alone will allow the legs to kick in alternation. Limbs which are grafted in the thoracic region of the spinal cord either become motionless or mere- ly exhibit slight twitches. The inability of the thoracic segment to co- ordinate limb movements is attributable to "faulty" neural connections 14 within itself, since the nuscles of the grafted limbs receive adequate innervation and normal nerve distribution from it (Detwiler, 1922; Piatt, 1956)- Armed with this kind of data, Straznicky (1963) took advantage of the fact that the avian pattern of locomotion is different from that of the much researched land tetrapod. By replacing the brachial segments with thoracic and lumbosacral segnents in chick embryos, he was able to confirm the results obtained by workers previously mentioned. Basically, he demonstrated that whereas wings innervated by the grafted thoracic «ments remained perfectly motionless, those innervated by the lumbosac- ral segments exhibited functional establishment. Such chicks, when walking, raised and lowered their rings at the shoulder Joint in parallel synchrony with the stepping of the ipsilateral leg. From all these results, one un- equivocal conclusion has been drawn: that clear differences prevail in the functional capacity of different spinal cord segments, and that these differences are determined already in early anbryonic life (Straznicky and Szekely, 1967). In the stuch' of coordination of limb movements, one of the inevita- ble questions is to determine mother a particular pattern is centrally programmed. The converse is to show if and how peripheral factors modify centrally occurring events. One of the most controversial studies has been the work of Weiss (1 9&1 a, b) in which the left and right forelimb rudiments of salamander embryos were interchanged at a stage when the an- tero-posterior axes of the limb had already been determined. When normal limbs developed, they were backwards instead of forewards. After nerve connections with periphery had been established, the grafted limbs moved Just as they would have done if they had been left in their oriynal 5 positions, working to move the animal backwards when the movements of the rest of the bow were working to move the animal forewards. Experi- ence of up to one year did not appear to modify the motor pattern. Be- cause such motor patterns develop in the absence of sensory innervation, their manifestation has been considered to be the consequences of exist- ing central programs. In a more recent paper, Szekely et a1. (1969) have recorded mscle potentials from eight muscles of the forelimb in freely moving, normal and deafferented newts. 'lhe myograms revealed delicate in- teraction of the antagonistic muscle groups. Despite a few irregularities in the placing of a limb, deafferentation of one or both limbs did not appear to alter normal activity patterns of the muscles. These investigat- ors, like Weiss, concluded that the brachial sement is capable of secur- ing tired activity of agonistic and antagonistic limb muscles without the necessity of receiving afferent feedback from the moving limb. Moreover, by employing Weiss's (1950) "deplantation" technique, Szekely (1967) was able to demonstrate that spinal segments from the limb level have the competence to coordinate limb movements even if they are isolated from the rest of the nervous system, and receive non-rhythmic excitation. The existence of central programs which determine the output pat- tern .of motor neurons exclusive of sensory feedback in vertebrates is supported by mam electrophysiological findings in invertebrate systems. In the locust flight system, the generation of normal output requires sensory input from peripheral receptors, but phasic information in these afferent signals is irrelevant (Wilson and Gettrup, 1963). The crucial point in this stuck lies in the fact that feedback, although present, does not determine the tang beat cycle; it simply serves to excite the central nervous system in a nonspecific way so that it operates faster. A similar 6 conclusion has been drawn for the sound-producing mechanism in the cicada (Hagiwara and Watanabe, 1956). Just as there are system that establish the existence of central programs, so can the peripheral patterning of motor output be shown. In the toad, Gray and Lissmann (191,6 a, b) demonstrated that the rhytlunical sequence of limb movements can only occur if at least one intact spinal nerve is present. Furthermore, peripheral influence is not simply a matter of local reflexes, but rather sensory input from each limb has some in- fluence on the posture of all limbs (Gray, 1950). In animals, deafferent- ation of even a single limb usually results in the failure of that limb to show locomotory activity (Lassek and Mayer, 1953). However, because severance of the sensory nerves in the distal portions of the cat's limbs does not prevent walking, it seems probable that the essential feedback comes from the proprioceptors in the proximal sections (Sherrinfion, 1910). In insects, too, removal of tibial and tarsal receptors from the farcral chordonotal organ in the locust results in changes in motor act- ivity and non-coordination of leg movements (Usherwood et al., 1968; Ush- erwood and Runion, 1970). Similarly, removal of coral hair plates results in overstepping (Wendler, 1966). In the cockroach, there is evidence to suggest that both central programming and reflex control are important in producing rhythmic movements of single legs and coordinating the move- ments of different legs (Pearson and lies, 1970; Deloonryn, 1971 ). The present stuck represents a departure from the observations of Hamburger and Oppenheim (1967) and Hamburger (1968), who reported that be- fore dey 17 of incubation, each limb in the chick embryo moves independ- ently of the other. After day 17, the two wings flap in unison, while the legs kick alternately. The questions I asked include: 7 1) From the earliest time in which an embryo can be safely opened and movements recorded, to the time of hatching, what is the temp- oral relationship between coordinated leg movements and total leg activity? In other words, does the degree of coordination increase over time? 2) If, in fact, there is a shift from relatively non-coordination to coordination, how does the change occur developmentally? Does it represent a gradual and sequential change or is it a sudden turn-on event? The studies carried out by Balaban and 11111 (1969) describe behaviors appearing around hatching time as being sudden changes in levels of performance. 3) In specific, does the change in the pattern of leg movements from relatively non- coordination to a high degree of coordination represent changes in local “events occurring at the lumbosacrel region alone, or do other parts of the central nervous system, es- pecially the brain, contribute in the development as well as in the actual performance of these coordinated movements? 14) Do sensory inputs modify the level and/or pattern of these move- ments? Table 1 illustrates the experimental design, dealing with the fol- lowing aspects: 1) The possible influence of the brain on the rates and pattern of coordinated movements of the leg. 2) The possible existence of a reciprocal influence between each half (right vs left) of the lumbosacral segment, and how this re- lationship affects both rate and pattern of coordinated leg move- ments . h) The possible existence of suprasegmental (brain) and segmental -x- (lumbosacral) interaction , and the degree to which this relation- ship influences rates and pattern of coordinated movements. In order to put these different aspects of the experimental design under test, experiments involving surgical manipulations of the central nervous system were carried out. The description of each manipulation is incorporated in the design, which also allows the following comparisons to be made: Table 1. Experimental Design. 1) Mid—thoracic transverse section (MTI'S) A, 2) Inmbosacral mid-saggital section (ISMSS) V 3) Inmbosacral deafferentation (mm) as h) mid-thoracic transverse section & (M'I'I‘S + 15163) Lumbosacral mid-saggital section a S) Mid-thoracic transverse section 8: (ms + ISDF) Imbosacral deaffe rentation «- ISMSS 4 6) MTI’S + LSFBS as ISIF 47)M'1TS+LSDF Control \)\'n\’ "D. .Ll. a N I. . In . u . . . n .a I a O I. n .v . . . MTS-I-LSDF M'fl‘S +ISI‘BS LSDF Schema of lesions . Figure A. MATERIALS AND METHODS All embryonated eggs were obtained from a flock of'flhite Leghorn chickens, which were raised by a commercial hatchery (Richard Hutting, Lansing). They were then stored in a refrigerator at 6 Celsius for a period of 2h-h8 hours in order to decrease the variability in the stages of development among individual embryos (Gottlieb, 1963). They were in- cubated on their sides (lateral position) at a temperature of 37.5 Celsi- us and relative humidity of 70-80 S in commercial incubators (Sears Roe- buck and Co., Mbdel No. 288.735). The incubators were not provided with contraptions for turning the eggs, and fur reasons stated in the surgical section of this study, no attempt was made to rotate the eggs manually. Preparation of Embryos To prepare embryos for surgical operations, eggs were removed after they had incubated for 2%r3 days. Each egg was then candled to mark the position of the embryo inside the shell. A sterilized probe was used to puncture the egg at the center of its blunt and over the air space in orb der to equalize the pressure inside the egg prior to carving a lateral window through which surgery is performed. Equalization of pressure pre- vents the embryo from adhering rigidly to the overlying shell membranes. The window itself was made by placing the egg on its side into a carved-out rubber egg-holder, whose receptacle is in turn egg-shaped. The portion of the shell to be removed was sterilized with 70 per cent ethyl alcohol. hath the egg securely placed in the egg-holder, a dental drill (Emesco Model No. 10), using round bits (Busch No. h and 5), was used to 10 11 dig an oval-shaped furrow on the shell. The circumscribed shell was then lifted with a pair of sterilized forceps. The shell membranes were care- fully peeled from the underlying embryonic membranes, thus creating a window, which exposes the embryo for manipulation. The drilling method offers two distinct advantages over other methods: it allows the experi- menter to accurately control the shape and size of the window; and it produces minimal disturbance to the integrity of the embryo even after extensive vascularization has occurred. In order to resolve the structures of the developing embryo in greater detail, i.e. neural from somites, vital stains were used. For all the experiments, neutral red embedded in an agar carrier was the pre- ferred stain. Staining itself consisted of dropping a small cake of the stain material onto the area to be operated upon, care being taken not to overstain. After adequate staining had been achieved, the cake was re- moved with a pair of forceps. Those embryos which were not sufficiently developed for operation had their windows temporarily closed with Scotch tape and were then returned to the incubator to be attended to later. Meal Operations Operations on embryos involving the creation of spinal "gaps" were performed by removing small segments of neural tube in the mid-thoracic region during stages 15-16 (Hamburger and Hamilton, 1951 ). The length of each gap was equivalent to the length of 2-3 somites. All other operations involving the sectioning of the lumbosacral cord in one way or another (mid-saggital section, and deafferentation) were performed during stages 17 and 18. A mid-saggital section of the lmnbosacral cord was favored by these later stages because of three main factors: during these stages, the neural tube has Just closed along the dorsal posterior portion, an event which enables the experimenter to make an accurate median section, 12 which divides the neural tube into equal left and right halves; the size and growth of the neural tube is of substantial magnitude so as to endow the tissue with inertia, against which the surgical knife acts to make a clean out without maceration; and lastly, these stages are imediately followed by growth and developnent of limb buds, whose increasing mass have a pulling effect on tissue surrounding each side of the cord, thus ensuring that the separated halves do not come together. This strategy obviated the placing of neural blocks, such as tantalum foil, along the mid-line of the neural tube . Operations which involved the removal of the dorsal half of the neural tube (deafferentation) were best performed if the neural tissue were first separated from the adhering pia and dura mater prior to the shaving of the entire lumbosacral segment. The operations described above made use of microsurgical technique whose basic elements and virtues have been described (Hamburger, 1960; Wenger, 1968). The vibrating needles used in these operations were driven by a phonograph crystal cartridge (Astatic Model N 14-2), a regulated pow- er amplifier (Elin RA-1100), and a wide range oscillator (Hewlett-Packard 200 CD). The exact shape and size of each needle were determined by the type of lesion to be performed. The basic hook-up equipnent is diagrammed below: 13 D. C. v--TCartridge I|||-J Oscillator A. be ° 7’ / line in B. Audio Amplifier Figure Be \\ surgical —) ‘ , knife \ ’I Block diagram of the microsurgical apparatus. The variable frequency oscillator was particularly suitable because at any given voltage, the madman amplitude of the vibrating needle tip depended on, among other things, the output frequency, the resonance char- acteristics of the crystal cartridge, as well as the material strength of the needle itself. It was generally found that for this combination of equipnent components, a frequency of 2300 Hz. , and an output voltage of 150-200 V were very optimal for most of the operations. when each operation was completed, the window on the egg was sealed with a cover glass, which was glued to the shell with a thin film of melt- ed paraffin. The egg was then returned to the incubator and left there 1h to incubate in the lateral position, with the cover glass uppermost. The eggs were not rotated during the remaining 152-17 days, even though ro- tating eggs enhances hatchibility. The reason for not rotating the eggs is based on the fact that, whenever young embryos are allowed to come in- to close contact with the cover glasses, they get stuck there, and invar- iably, death ensues. Recoflg Enbryos were removed from the incubator on day 19 (at 156 hours) of incubation. Only those embryos whose beaks had penetrated the chorioallon- toic mmbrane and were exhibiting strong pulmonary respiration were used. Each embryo had part of its shell removed from the blunt and so as to ex- pose the head and wing areas. The chorioallontoic membrane was caremlly peeled in the direction of the blood flow in order to avoid excessive bleeding in the still existing chorioallontoic blood vessels. The head, which is normally tucked under the right wing, was pulled out and the coiled neck straightened. The two wings were placed in such a way that they drooped over the shell on each side. Similarly, the neck and head were allowed to droop forward over the shell in a position tangential to the wings. This symetry was crucial to the proper recording of leg move- ments in that it kept the movements of the other parts of the embryo's W (wings, head) from pulling on or interfering with the strings at- tached between the feet of the preparation and the transducers. With the aid of a blunt pair of forceps, each leg was pulled up and out of the shell; a piece of thin thread was attached to the web of each foot between the second and third digits. The legs were then gently pushed back to their normal position within the shell. Each embryo was then placed in a glass fish-bowl which, in turn, was partially suhnerged in a water bath 15 (Precision Scientific 00., Thelco Model No. 83), where temperature and relative humidity were similar to those of the incubator. The loose ends of the thread were attached to force-displacement transducers (Grass type FT .03). Care was taken to ensure that tension on each thread was neither too taut so as to hinder leg movements, nor too loose, such that some moments went unrecorded. The transducers were connected by means of cables to a polygraph (Grass Model 8:3?) using D.C. pre-amplifiers. Record- ing of leg movements was commenced after the animal had adapted to the new situation for a period of two hours. Drring this time also, the mach- ins was turned on and the amplifiers calibrated. Measurements From the polygraph print-out, two basic measures were selected and analyzed to account for the ways in which leg movements occur in the pre- hatched chick: frequency and pattern. Needless to say, frequency in this study refers to the number of coordinated leg movements per unit time. Two basic patterns of coordination were found in the course of experiment- ation: the simultaneous pattern in which the two legs move in unison, and the alternate pattern in which the two legs move within a determined time unit of each other. A criterion established to distinguish one pattern from the other was that, in order for any movements by the two legs to be considered simultaneous, they had to occur within one second of each oth- er. For the alternate pattern, the movements had to occur between 1-2 sec- onds of each other. By this definition, then, whenever movements by the two legs were separated by an interval of more than 2 seconds, they would be considered uncoordinated. In reality, however, non-coordination (re- gardless of the pattern of movements displayed) consisted of movement(s) made by one leg, unaccompanied by the corresponding movement(s) of the other leg. RESULTS In order to determine whether or not coordination of leg movements incorporates developmental features during the last two days of incuba- tion (from day 19 to hatching), two linear scales have been used: the rate of coordinated movements as percentage of total activity (coordinated plus uncoordinated movements), and absolute rate. Figures 1-5 illustrate that coordinated leg movements, regardless of the pattern displayed, constitute about 90 per cent or more of total activity. This ratio does not alter in am' significant degree during the two-day period. These results also hold true for the absolute rate; the quasi increase in frequency of movements observed in the two groups (fig- ures h and 5) is not significantly different from the zero slope (regres- sion analysis). Figures 1-5: the top line represents coordinated movements and its scale is the left vertical axis. The bottom line represents frequency and its scale. is the right vertical axis. The legend for each group is provid- ed below each figure. 16 17 100- . ,\ ~1o.o \./.‘_.’___.\.v___,o_—g——t\./ e 90.1 " 900 ’8 80‘ h 8.0 g 70" ‘" 7.0 +3 5 60" L 600 E 2 goJ .. s.ov ‘5’ h0« \. . h.0 g \ .’ ’,e\\ e ------ .",e-—-e‘ ‘ ‘ O" 30. \\ / ‘0s 1” . \. I. 3'0 a 0’ ~"’ I“ 20‘ b 2.0 10-1 - 1.0 E 8 12 15 25 I h ’3 12 16 20 1 19 20 21 Days and Hours of Incubation Figure 1. Control. 1mq -1000 K.\./. . .\./O\ /. 901 \,/-/ ' - 9.0 80" " 800 g 70' " 700g 0 60" P 6.0 E 3 50" b 500V h [‘0- - boo § 30-! ” \\ .‘ /.\ r 3.0 g ’1 o-——.‘ [I “O-_~.- ’, ‘\ ”'. k. 20- ' ‘°’ ‘“" " P 2.0 101 "' 1.0 . m a PE 16 76 a u 8 T2 T6 356—7" 19 20 21 Days and Hours of Incubation Figure 2. Mid-thoracic transverse section. 100- : ; : _ . -10.0 . :‘./ \._——O——-——'.N. 9o— ) 9.0 80- 1' 800’s 3 70" ' 700.3 60- ’I.“‘. r 6.0 l’.‘\ ll \\ g 50" I’,o’ ~‘~.‘~‘.\ [’0 \\\ ”’. r 500V 140- . \‘ ’ . P hoot: a 30- - 3.0 2 k. 20. . 2.0 10.4 " 1.0 19 20 Days and Hours of Incubation Figure 3. lumbosacral deafferentation. 1m1 .K°—~e/.\/.\°\./.\O\. .1000 90- P 900 80-1 "' 8.0"? E 70‘ ’;\\ ,’.\\ - 7.0g 60< ,' \ X'X ,’ x. - 6.0 g I / \ / \./ \\ I. \\ 8 SM I ‘ / \, - 5e0v ,0, ‘ I, 3 140-1 [0” \.l P 14.0 ‘3 I 30" I, " 3e0§ ’ In 20" " 200 10- - 1 .0 1; b #2 W l h 8W: ' 19 20 21 buys and Hours of Incubation Figure 1;. Lumbosacral mid-saggital section. W «I 3W 21 30-I 20- 1 0-1 19 Figure 3. w 30- 20- 1 0- m .x. ~./.\./.\o\./O\.\. 18 26 I '“E 20 Days and Hours of Incubation 1 6 8 12 16 21 lumbosacral deafferentation . 3‘./.\._____...__———.~. 1 h1000 I 9.0 A - 8.0“ E " 7e0:3I mg "’ 5003 I- h.0§ a b 300 8 Ch " 200 " 100 E3 19 Figure h. 17-13—71? I 20 Ihys and Hours of Incubation Lumbosacral nid-saggital section. EB1§16§6I’ 21 19 1m. fix. .\. . . ./O—-—.~. . '10 .0 901 *- 9.0 80, b 8.0“ /°\ a 7d [I] \\ b 700% I. \ m 60" I \\.’,”.‘~‘. '- 6.05 z I, \\ 50" l.’ \O\‘\. I— 5.0é ho. - no? 0- --—e’ g 304 3'08 In 20" "' 2.0 1m ‘ 1'0 I—E_'8 12 1F 2'6 I' w11*1 12 16 TIT—I 19 20 21 Days and Hours of Incubation Figure 5. lad-thoracic transverse + Lumbosacral mid-saggital sections. ActivitLPeak as a Basis for Analysis When one draws a curve which describes an activity of an animal re- corded over a period of time, it is observed that the temporal pattern is such that each animal builds its activity to a peak, after which activ- ity decreases. Both the build-up and decline phases may be gradual or pre- cipitous. According to this study, there occurs only one peak per day, al- though the time of the appearance of the peak seems to be related neither to a strict chronologically-detemined developmental stage, nor to the time of the day. Nanetheless, this build-up of activity to a peak is one characteristic that is shared by all animals. Therefore, activity peak has been selected as a reference point at which the axis of one curve is superimposed on another in order to compute average activity level. It has also been selected as a single point at which the level of activity 20 of one group is compared to another. In doing so, developmental time in terms of days and hours of incubation has been collapsed into a single curve, with two temporal components: time before and time after activity peak. To make comparisons on the rates of coordinated movements among the various groups, the Mann-Whitney U tests were chosen. The experiment- al design (Table 1) was adopted in order to make comparisons (activity peak curves) which provide answers to the following questions: 1) What kind of influence, if any, does the brain have on the fre- quency and pattern of coordinated leg movements? 2) Are there influences within the lumbosacral spinal cord which affect both frequency and pattern of leg movements? 3) Does sensory feedback play a role in the degree to which the frequency and pattern of coordinated movements occur? lI-7) Is it possible to show the existence and magnitude of the brain/ lumbosacml interaction, insofar as it affects both frequency and pattern of coordinated leg movements? I. Effects of mid-thoracic transverse section on the fragrancy and pattern Wordinated movements. When the frequency of coordinated movements in animals with mid- thoracic spinal gaps is compared with that of the controls, a significant difference is obtained (0.05 >P >0.02, two-tailed; Fig. 6). In conform- ity with the rest of the groups, comparisons are made at the activity peak of each group. 21 Frequency (movements/min) 10.0q 9'01 K- Control 8.04 ‘ ---- ° MTTS 1.0- 6.0~ 5.0- . 14.04 /\ 3.0- /. ’l’ ‘~\::\.~—. 2.0. ,:./_, , ”I ‘ “v ~ - .\\. 1 .O~ 3/. \\‘. 16‘ 8 6 I; T—F—‘f‘ 11 6 T—"I'o' Time in Hours From Peak Figure 6. The pattern of coordinated movements in animals with spinal gaps of the thoracic segnent is basically one of alternation. In this regard, the spinal animals were no different from the controls. Upon hatching, the chicks were able to tuck their legs under the venter and perform the right- ing response when placed on their backs as well. These spinal animals, however, unlike the controls, could neither stand nor 1311:. The only loco- motory moverimnts they were capable of were those akin to crawling, which was achieved by rapid kicking of the legs braced against a rough surface. II . Effects of lumbosacral mid-saggital section on the frequency and pat- m coordinated movements . Animals with, essentially, separate left and right halves of the lumbosacral spinal segnent have significantly higher rates of coordinated activity than the controls (.01 > P > .005, two-tailed; Fig. 7). 22 10.0- 9 .0. -—-—-—. Control 8.0- -- —-- - Ierss Frequency (movements/min) 1 .0-1 ./ 16—8 6 14723571: 6 8 10 Time in Hours From Peak Figure 7. The pattern is different from that of the two previous groups in that simultaneous kicking of the legs, rather than the alternate config- uration, is observed. Yet the rapid flexions and extensions of the legs in unison are so precise that they must be reckoned as coordinated move- ments. As with all other groups, uncoordinated activity consisted of sin- gle or multiple movements by one leg, unaccompanied by corresponding move- ments of the other leg. This type of lesion imposes a severe posture on the animal after hatching. The newly-hatched chick lies prostrated on its venter, while both legs are rigidly extended parallel to each other he- hind the rest of the body. Any attempt to forcibly flex and tuck the legs under the animal's venter is met with persistent resistance, followed by a resumption of the extended-legs posture. These animals cannot right themselves, or do so with extreme difficulty. 23 III . Effects of deafferentation of the lumbosacral segment on the frequen- gand pattern of coordinated movements. Removal of the dorsal half of the spinal cord in the lumbosacral segment constitutes a form of deafferentation. Animals with such lesions (ISIE‘) mnifest a high level of coordinated activity that is significant- ly different from that of the control group (0.05 > P > 0.025, two-tailed; Fig. 8). Deafferented animals perform simultaneous movements of paired limbs in a fashion that is indistinguishable from that of mid-saggitally dissect- ed preparations. Similarly, the posture assumed immediately after hatching is one in which the chicks prostrate on their venters, with both legs ex- tended behind the body. 1000‘ 9.0.1 '°———°Control a 8e0‘ . ----- .I‘SDF \ u a 7.0- E 6004 ” \‘\ 8 ’ \ V 5.0‘ [I ‘0‘ g “'°‘ '"-"’ /\ 3.0-1 . . ‘.“‘~e h / \' ° 200‘ ./ \ 100‘ I/ If B 6 u M? I; 6 8 16’ Time in Hours From Peak Figure 8. 2h IV. Effects of mid-thoracic transverse and lumbosacral mid-saggital sec- tions (m3 + 1.81485) on the frequency and pattern of coordinated movements. Preparations with both transverse and longitudinal sections not only exhibited rates of activity which were significantly higher than those of the controls (P 4 0.002, two-tailed; Fig. 9), but manifested patterns of activity together rdth postures which parallel those of the mid-saggitally lesioned animals as well. 1000‘ 0 Control 9.0-1 ’ \ I’ \\ °--- --- M'I'I‘S + LSNSS 8.0J * , \ 700‘ I \ 600‘ 1’. \ 1400“ I” /\ \O\ I e—--" \\\. 3.04 /' \._. 2 .04 ./. \. 10 8 6 h 275 2 h 68 10 Time inHours From Peak Figure 9. A description of coordinated movements accompanied by changes in pos- ture after hatching in the double-lesioned animals fits quite well, with minor variations, that of the animals with longitudinal sections alone . Thus, simultaneous kicking of the legs and extension of the legs behind the body are consistently observed. One of the postural differences that 25 exists between a double-lesioned animal and one with a.mid-saggital sec- tion is that the degree to which the legs are rigidly extended behind the body subjectively appears to be less in the former than in the latter. The reduction in rigidity (measured by flexing the legs ventrad) is, perb haps, attributable to the thoracic transection. V. Effects of deafferentation and mid-thoracic transverse section on the frequency and pattern of coordinated movements. If in addition to deafferentation, embryos are subjected to mid- thoracic transactions, a higher rate of activity is observed in the doub- le lesioned animals as compared to the controls (P 4.0.002, two-tailed; Fig. 10). The purpose of the transverse section is, as it was for group IV (MTTS + LSMSS), to prevent the influence of the brain from reaching and acting upon the lumbosacral segment of the spinal cord. 1"0' X“ I /\ \\ / \ 300‘ e, . \. . 2.0. .'// \ I... / 1000- 9 O- . . COntI‘Ol A / \ a 8.0- , \\ .--__. m'rs + LSDF } ,’ ‘ g 7'0“ [I \\ E 600'! I: \.\\ ‘3’ 5.0! I. I \‘eo" ’ .\ E Time in Hours From.Peak Figure 10. 26 The pattern of‘movements in the double lesioned animals conforms to that of groups II, III, and IV, i.e. the simultaneous kicking of the legs. 0n the other hand, unlike group IV (MTTS + ISIBS), in which the post-hatch posture exhibits a marked reduction in the degree to which the two legs are extended behind the body, presumably because of the transverse section, the same transaction bears no effect on the posture of the deaf- ferented embryo. That is to say, the rigidity of the legs in the deafferb ented embryo is subjectively equivalent to that of the deafferented and transected embryo. .Lgmhgsacral,mid-saggital section vs. mid-thoracic transverse + lumbo- sacral mid-saggital sections (LSMSS vs MTTS + LSMSS). The purpose in comparing these two groups is to determine if the level of activity of the legs is influenced in any significant degree by the interaction of the brain and the lumbosacral spinal cord. Suffice it to mention that the two groups share two other characteristics which make this comparison even more valid: they exhibit the same pattern of coordin- ation; and they both share the mid-saggital section. Figure 11 illustrates that animals with both mid-thoracic transections and longitudinal sections exhibit higher levels of leg activity than those with longitudinal sec- tions alone (0.05 >P > 0.025. one-tailed)- 27 1000'! '-——-o LSYSS 900‘ ’5 8 0--—-e FITS 4" LSPBS .0 ’ E -I O 1E} 700‘? g 600‘ 3 V 500‘ \ 8‘ x g 1400" \.\‘ 8- x, a 3e0" (I. 2.0J 1 .0- 1O 8 5 h 2 F 2 74 6 8 10 Time in Hours From Peak Figure 11 . VII. Deafferentation vs deafferentation + mid-thoracic transverse section. These two groups are compared in order to assess the relative mag- nitude of the influence of the brain on the lumbosacral region. Figure 12 shows that the double lesioned embryos show a greater level of activity than those with simple deafferentation (0.025 > P > 0.01 , one-tailed). Frequency (movements/min) 10.0- 9.0- 8.0- 7.0- 6.0- 5.0- Figure 12. 28 6 11 2 P 2 11 6 8 Time in Hours From Peak 29 3mm” 3mm" oaman awed he mama awed Ac mmoa Ac camah mmoH Ac eoesofio teases eoepofio 823%. e663 mmoH An aon An mon An awed An awed An masseuse o: A6 meanness on A“ a As suspense o: A6 masseuse Au mmppmom moccadpflesam nsoonmadafifim msoosmpHeEAm uncommpH52Hm opennopam zmmaeHao¢ mama + webs. mmzmg + mesa anmq mmzmg was: nofimoa no make . 3.93m no Nude-Em .N canes DISCUSSION As re-statement of the problem to which the present study addressed itself, a search was made for the development of coordinated leg movements in the pro-hatched chick. The temporal relationship between coordinated activity and total leg activity envisaged changes in pattern as well as in magnitude. Finally, through the various manipulations of the central nervous system, we sought to relate behavioral patterns and their'magni- tude to their'underlying neurological substrates. Temporal Pattern of Coordinated M0vements. While this study failed to establish the existence and.mode of de- velopment for coordinated movements, there were two fundamental ration- ales for attempting to do so. First, since coordination of'limbs does not set in until day 17 of incubation (Hamburger and Oppenheim, 1967), it was reasonable to assume that some residual component of this development could still be shown between day 19 of incubation and hatching. The sec- ond rationale has been proposed or alluded to for teleological reasons (Hamburger and Oppenheim, 1967; 0ppenheim.and Nareyanan, 1968). That is to say, a high degree of coordination, in terms of absolute rate and types of'movements, is thought to be necessary in order to enable the an- imal to escape from.the mechanical confines of the shell. That the rate and pattern of coordinated movements do not alter during the last two days of incubation has been adequately shown in this study (figures 1-5). Another aspect of the ontogeny of’coordinated movements, the exist- ence for which could not be established in this study, is the mode or pattern of development . By pattern of development it is meant, whether a 30 31 high degree of limb coordination such as is observed in a 19-day-old embryo results from gradual and sequential changes in central and/or per- ipheral factors, or whether it is simply a function of sudden, turn-on events. Kuo (1967) has argued in favor of the former as the mechanism un- derlying development of most behaviors. Yet it is known that certain be- haviors such as vocalization, head-lifting, and eye-opening in chick em- bryos appear very suddenly, and are thereafter maintained at high levels (Balaban and Hill, 1969). Similarly, the back thrusts of the head in duck embryos suddenly increase in frequency about 16 hours prior to hatching (Oppenheim, 1970). The righting response and, to a large degree, hatch- ing belong to the latter category. It is evident from this study that any further attempts to stuw the ontogenetic facets of limb coordination must be directed to embryos which are at least younger than 19 days, a task which imposes severe limitations on the ability to expose the embryo well enough for observations or recording and still maintain it alive. A temporal pattern in the form of a slight increase in the rate of activity was exhibited by two experimental groups (ISMSS, MTI‘S + 1.8163; figures 14 and 5). The trend toward an increase is attributable to the mid-saggital section since other groups with other types of lesions, to- gether with the control group, did not show such a pattern. It is believed that a larger sample size in each of the two groups might have rendered the increase significant. Were that to occur, such an increase would be in contradiction to the normal activity pattern in which there is a pre- cipitous decline in overall activity after day 17 of incubation (Hamburger et al., 1965; Hamburger and Oppenheim, 1967). This decline has been shown inthis studytohavereachedanasymptotebyday19 andremainsinthat level until about the time of hatching. At any rate, further questions 32 could be asked about the nature of the increase; whether it is the pro- duct of a change localized in the lumbosacral region, or whether an early mid-saggital section somehow activates the organism to perform a higher level of overall activity, and not just the legs. It would also suggest that the general decline in overall activity observed in normal embryos could be partially reversed by certain specific lesions. That in turn would lead one to ask the question as to whether or not the onset of integration of moments and the decline in overall activity seen shortly after day 17 have a cause-effect relationship. By continuous recording of leg momnts, it has been possible to formulate a description of the temporal pattern of activity, which arlmin- ates in a peak, wherein activity level often represents a three-fold in- crease from a "base-line" level. To In kmwledge, such a phenomenon has not been described before for chick or duck unbryos, either as a daily- occurring event or as an ocarrrence specific to animals in their last days of incubation. Activity peak of this nature has eluded observance and, therefore, description primarily because most investigators use a stand- ard 15-mimrte observation period or a variation thereof (Hamburger and Balaban, 1963; Hamburger et al., 1966; Decker and Hamburger, 1967). This sampling approach makes it difficult to reconstruct a relatively short- lived event such as peaking, especially since under standard incubation conditions, the time of appearance of activity peak seems to bear no re- lationship either to a strictly develomental stage or to the time of the day. Effects of Mid-thoracic Transverse Section on the Fmggemd Pattern of Coordinated Hovanents. The rate of coordinated movements was significantly lower in animals with thoracic gaps than in the controls (P 4 .05). These results are in 33 agreement with the early work of Hamburger et al. (1965), who found that spinal embryos between the age of 8-17 days exhibited significantly low- er amounts of leg motility than controls. In disagreement with both of these findings is the study made by Oppenheim and Narayanan (1968), who failed to show such a difference in 19-20-day embryos. Vhat the present stucw points out, contrary to Oppenheim's and Narayanan's implicit sug- gestion, is that the lowering of leg activity in spinal animals is not consequential to the 8-17-day embryos only, but attends to older embryos as well, Hamburger and co-workers have attributed the low level of leg motil- ity in spinal animals to the onset of degeneration of the spinal cord mo- tor neurons. Two arguments are presented here to suggest that additional factors other than degeneration might be responsible. First, if low lev- el of leg motility were the result of degeneration, then, assuming equal or equivalent activity, older embryos which would be attended by extens- ive amount of degeneration should perform movements at significantly low- er levels than younger embryos (i.e. comparing rates in 19 and 20—day em- bryos or Hamburger's 16 and 17-day embryos). This is not what is observ- ed. Second, in this stuchr, where extensive lesioning was employed (multi- ple lesions, figures 9 and 10), a corresponding lowering of activity rep- resenting extensive degeneration should be expected. In general, however, the results have shown Just the opposite. The capacity of the lumbosacral segnent of the spinal cord to per- form alternate coordinated movements of the legs without the aid of the other parts of the central nervous system has been borne out in this stuchr as well as in other studies (Hamburger and Balaban, 1963; Oppenheim and Narayanan, 1968; Detwdler, 1936; Szekely, 1963, 1967). Similarly, locomo- tion in mantids (Roeder, 1963) and in crickets (Huber, 1959) appears to 3h be highly organized in the thoracic ganglia. The role of the brain is seen as one of providing stimulus for excitation and inhibition to the otherwise competent thoracic ganglia. Furthermore, even highly complex behaviors such as flight and jumping in locusts can be accomplished af- ter complete decapitation (removing the subesophageal ganglia) (Wilson, 1961 ). A few studies, however, emphasize the importance of the higher cen- ters in the actual execution of movement patterns. Thus, Decker and Ham- burger (1967), Hamburger and Narayanan (1969) suggested that the perform- ance of coordinated movements such as wing flapping and tucking of the head under the right wing in chick embryos is correlated with the matur- ation of higher centers. Hughes and Prestige (1967) found that tadpoles with spinal transections of the cervical region fail to exhibit swimming movements of the legs. Effects of lumbosacral Mid-saggital Section on the Frequency and Pattern moordinated Movements . Animals with longitudinal cuts performed simultaneous coordinated movements at higher rates than the controls (P 4 .01 ). These results point out that the two halves of the lumbosacral segment, in addition to providing reciprocal inhibition for the execution of some reflexes, as is commonly found in vertebrates, there is also mutual inhibition of en- dogenous activity. An analogous situation has been reported for the meta- thoracic ganglion of the cockroach, in which it was fmmd that activity of marv motor neurons was partially suppressed by neural connections with the opposite side of the same ganglion (Weiant, 1956). That the two legs should perform simultaneous movements after neur- al connections between the two segments of the lumbosacral region have 35 been severed is an unexpected finding. The type of information which the isolated segments use to signal the movement of their respective legs in synchrony is not yet understood. Since mrther isolation of the segments from the possible influence of the brain (mid-thoracic transaction) re- sults in the same basic pattern, the brain could not be operative in providing the necessary information. The postural reflexes shown irmnediately after hatching indicate that the severing of cross-fibers between the two segments so strongly activ- ates the extensor muscles that the opposing action of the flexors is over- come. The net result of this action is the rigid extension of the lags behind the body, a phenomenon somewhat akin to mid-collicular decerebra— tion (Sherrington, 1906). Effects of Mid-thoracic Transverse and Lumbosacral Mid-saggital Sections on the Frequency and Pattern of Coordinated Movements. The two limbs in the double-lesioned preparation move in perfect synchrony (simultaneous) at a much higher rate than that of the control (P < .002). Furthermore, the frequency of movements in animals with both the transverse and longitudinal cuts is greater than that of animals with the longitudinal alone. From these results it can be inferred that the ' nervous system anterior to the thoracic region normally exerts an inhibi-e tory influence on the lumbosacral region, but that this influence is best demonstrated after the integrity of the lumbosacral segment has been dis- rupted (mid-saggital section). The literature is abundant with studies showing inhibition of the spinal cord or its analogue by the brain (Road- er, 1963; Huber, 1959; Decker, 1968; Weiant, 1956). Roeder has reported on the effect of severing the connections between the brain and the sube- sophageal ganglion in the mantid as being continuous forward locomotion. 36 A similar situation obtains for the cricket except that the locomotor mechanisms in the thoracic ganglia are activated by the central bochr, in addition to the subesophageal ganglion, and that both are in turn regulat- ed through inhibition arising from the mushroom bodies (Huber, 1965). In the chick embryo, Decker has shown that unilateral or bilateral extirpa- tion of the otocysts results in increased total body motility, suggesting that one of the normal functions of the vestibular centers is to check the spontaneous activity of the spinal cord motor neurons. Weiant's study, which has alreach' been cited in the previous section, revealed that act- ivity in the efferent fibers of the metathoracic ganglion was greatly en- hanced as soon as the ganglion was further isolated from other nerve con- nections. The greater the isolation, the higher the frequency of firing. An alternative explanatial which accounts for the tremendous in- crease in the rate of activity in the double lesioned embryos is to pos- tulate the existence of a mechanism by which the brain inhibits some neurons in the lumbosacral cord. These neurons, in turn, inhibit the mo- tor neurons involved in the performance of leg movements. Such a mechan- ism could be said to operate through inhibition of inhibition. Such a mechanism would also account for the lowering of activity observed in embryos with mid-thoracic transections. The validity of the mechanism is even more apparent if it is assumed that the lumbosacral inhibition alone produces greater effects than that which is projected from the brain. Figure 0 illustrates how this mechanism might operate: 37 BRAIN I= Inhibition I LUMBOSACRAL cons v, Figure C. Block diagram of the inhibitory mechanism. Similarities between animals rdth saggital sections and those with saggital plus transverse sections have been mentioned alreacw; similari- ties in elevated rates of activity, of posture, pattern of movements, and or the temporal pattern of activity during the last two days of incubation. 01' these, perhaps the most intriguing is the pattern of movements. The very reason for carrying out double lesions was to test the hypothesis as to whether or not the animal could and might be using information from the brain to coordinate its movements. That that is not the case has been shown already. A few spearlative explanations are offered to account for the simultaneous pattern of movements. The first is one of mechanical fac- tors; that is, because recordings of the leg movements are made while the animal" in still partly propped up in the shell, the tarsals and tarsal Joints are frequently braced against the shell. The mechanical force re- sulting from the two legs braced against the shell, and from the friction arising from the movement as the tarsals scrape the inside of the shell--- 38 these factors could simultaneously excite the afferent fibers in the two segments, and the afferents, in turn, send volleys to their respective motor neuronal pools, and thus enable the legs to move in phase. Second, proprioception as the source which signals the next sequence of movements camot be ruled out , especially if severing of the connections between the two segments favors the development of biological oscillators in each segment. These oscillators could be synchronized by the position of the limbs, or by the velocity of the limb movements themselves. A third and remote possibility is for the accumulation of a large resevoir of transmitter substance in between the two half-segments. Were the substance to possess the property of exciting the neurons embedded along the inside borders of the segments, it is conceivable that tonic stimulation of the motor neurons might take place. The simultaneous pattern of movements makes it possible for a test- able statement to be made about the normal ontogeny of coordination. Name- ly, that the onset of alternate coordinated movements does not simply or- iginate because of the establishment of competent structural and function- al connections between the two halves of the lumbosacral segment, but al- so because of a host of other changes which take place in other areas of the lumbosacral region and even beyond. To put it another way, if one were to observe the movements of a mid-saggitally sectioned preparation before day 17 of incubation, it is predicted that the animal will exhibit the random and uncoordinated movements of the two limbs to the same degree that one sees in the unoperated control. Furthermore, if it is assumed or shown that the developnent of the alternate pattern i.e. both types of patterns begin to operate around day 17, then it can be further postu- lated that both of these patterns are subserved by common features which ~|.-. erg; 39 undergo changes to trigger, or allow, coordination to take place. The difference between the two patterns, therefore, would be reduced to the mere presence or absence of cross fibers. Effects of Deafferentation of the Lumbosacral Region on the FrequencL and Pattern of Coordinated Movements. The rate of coordinated movements in the deafferented animals was significantly higher than in the controls (P < .05). One characteristic which is unique to the deafferented animal is uniformity of activity. Ev- en though it shows activity peak as such, the element of dynamic range (high activity interspersed with low activity), which is so prevalent in all other groups, is at a minimum or lacking. It is as if the interneurons and the motor neurons associated with the initiation and maintenance of spontaneous activity are firing at a fairly constant rate, which in itself would suggest that one of the functions of the sensory cells is to check endogenous activity. That sensory input determines motor output, partic- ularly in terms of reflex activity control mechanism, is cornmon knowledge; but data are scanty to support the notion of long-term suppression of free-running motor activity by sensory input. Bud: (1961 ) succeeded in demonstrating that endogenous activity originating in the terminal process- es of large axons that form the ocellar nerve in the insect is inhibited by light impinging on the photoreceptor cells. The neural signal transmit- ted to the brain, therefore, represents a compromise between endogenous activity of the fibers and the inhibition imposed by the photoreceptors. Decker's (1968) experiments in which the otocysts were removed, led him to conclude that vestibular mechanisms inhibit activity in the neurons of the spinal cord. Although sensory organs and their central connections have not been implicated in this study, it is worth mentioning that exper- iments in which the influence of the brain was excluded from reaching the ho deafferented lumbosacral cord showed increased activity of the legs over those in which only the lumbosacral cord is deafferented (P 4 0.025). In conflict with these results is the study carried out by Hamburg- er et a1. (1966), who reported that extensive deafferentation of the chick embryo neural tube from the lumbosacral region all the way to the thoracic area does not produce a marked effect on the motility of the legs in embryos that were between 8% and 15 days of incubation. They re- ported, however, a sharp decline in leg motility in 17-day embryos, an event attributed to numerical depletion and partial degeneration of the lateral motor columns. In yet another set of experiments, Hamburger and Narayanan (1969) deafferented the trigeminal area of the head, and then proceeded to record total body'motility in the usua1.manneru They found no significant differences between normal and deafferented embryos up to 15 days of incubation. It is difficult to reconcile the differences be- tween their results and those reported in this thesis, except to mention that their studies were carried out during the early and middle life of the embryos, while those reported here were conducted during the termin- al portion of the incubation period. The pattern of movements observed in the deafferented animals is the simultaneous type. The loss of the alternate pattern, or rather'lack of its developnent following deafferentation, can be attributed to two things: the sensory cells and their connections in the central nervous system.are instrumental to or'mandatory in the development and execution of the alternate pattern,:much as they are important in the perfonmance of crossed extensor reflexes; and that shaving the dorsal half of the spinal cord removes cells which nonmally tonically facilitate the inhib- itory coupling between motor centers on the two sides of the cord. The )41 letter, of course, would also serve to explain why there was increased activity in the deafferented group. khich of the two mechanisms is oper- ating ranains to be worked out. The importance of sensory feedback in the control of rhythmic move- ments in many systems has been emphasized. Huber (1960) reported that sound production by crickets is a fairly complex behavior, involving feed- back from the nnrsculature. In the locust, removal of tibial and tarsal receptors results in changes in motor activity and uncoordinated leg move- ments (Usherwood, Runion, and Campbell, 1968). Overstepping in the stick- bug occurs after the removal of coxal hair plates (Wendler, 1966). The cockroach displays such diverse locomotory patterns that investigators were led to conclude that such a repertoire could only exist if the nerv- ous system possessed sufficient plasticity so as to meet various contin- gencies (Bethe, 1931). Despite the existence of demonstrably rigid central programs which form the basis of cockroach locomotion (Hughes, 1957), pro— prioceptive reflexes with both tonic and phasic components are superim- posed upon these central programs (Wilson, 1966; Pearson, 1972). In this study, the simultaneous pattern of movements, which persists after complete deafferentation, indicates that for the two legs to move in unison, they must receive simultaneous comnands. There exists ample evidence to suggest that central factors alone can account for some pat- terns of movements. Neither the overt rhythmic leg movements in the milk- weed bug nor the underlying activity burst in leg motor neurons is affect- ed by deafferentation (Roy and Wilson, 1969). It has also been establish- ed that the rhythmic discharge of the swimret movements of the crayfish does not require sensory feedback (Ikeda and Wiersma, 1961;). The abdominal ganglia are quite sufficient in initiating and perpetuating the rhythmic ’42 output. Even when the abdominal nerve was completely isolated, rhythmic- al bursts of electrical activity in the motor roots leading to the swim-- merets could still be recorded. Another example of very rigid central patterning is the sequence of motor impulses from the cerebral ganglion of the bivalve mollusc Liza, which normally brings about the retraction of the mantle and closure of the shell (Horridge, 1961). Single stimuli applied to the preganglionic (sensory) axon is capable of eliciting a se- quence of motor impulses from the ganglion in which up to 10 motor axons can be individually identified; the patterning of the impulses being sim- ilar to successive repetitions. The sequence does not change when all oth- er nerves to the cerebral ganglion are severed and when the motor nerve is out beyond the recording electrode. The sequence does not depend on proprioceptive feedback from the movement it causes either. Web-building movements in the spider are not abolished by amputating one or two legs (Szlep, 1952). Von Holst (1939) described the "superposition" effect in the teleost, whereby the amplitude of one of two fine beating out of phase with each other is determined (may be increased or decreased) by its phase relationship with the beat of the other fin. This superposition effect persists in the deafferented spinal fish. The scratch reflex of the dog consists of a fairly complex movement, requiring the participation of 19 muscles (Sherrington, 1931 ); yet it can be elicited by a simple stimula- tion from the deafferented hindlimb. It has also been shown that deaffer- entation of both hindlimbs in the dog does not abolish rhythmic movements during normal walking (Tourita, 1967). Engberg and lundberg (1962) record- ed EMG from the extensor of the knee (m lateralis) in the cat; they then correlated the EMG with limb movements photographed at high speed. They discovered that a burst of activity of the muscle appeared even 143 before the foot had had a chance to touch the ground; that is to say, before the muscle was stretched by the weight of the body. In most mam- mals, swallowing is a complex behavior in which activity in the motor neurons scattered all the way from the mesencephalic level to the third cervical level is integrated to produce a coordinated pattern of contract- ion in about twenty muscles (Doty and Bosma, 1956). Once swallowing has been initiated, the temporal pattern, duration, and amplitude of contract- ions are independent of sensory feedback. Finally, the work of Szekely (1969) demonstrated that deafferentation of one or both forelimbs in the newt does not alter normal activity patterns in the agonistic and antag- onistic muscle groups in unrestrained animals during normal walking. Gen- erally, a system which continues to maintain its basic output pattern in the absence of sensory feedback is thought to be under the control of pacemaker mechanisms. SUPMARY OF AND CONCLUSIONS FROM LESION EXPERIMWTS The results obtained from the various manipulations of the central nervous system make it possible to outline ways in which various parts of the nervous system interact to influence both activity level and patterns of leg movements. From the outset, it ought to be pointed out that the only form of influence demonstrated by the methods used in this study is one of inhibition. Frequency of Coordinated Movements. Even though a decline in activity was observed in embryos following mid-thoracic transections (Fig. 6), the influence of the brain on the lum- bosacral cord is considered to be one of inhibition rather than excitation. Evidence for this conclusion is derived from the demonstration that embry- os with double lesions (MT'I‘S 4- LSFBS; ms + ISDF, Figs. 11, 12) perform movements at rates higher than those of the corresponding single lesions. The apparent disparity between the two opposed findings lies in the belief that the inhibitory effect of the brain markedly manifests itself only when the integrity of the lumbosacral region is disrupted (ISMSS, ISDF). In other words, an integral lmnbosacral cord is capable of absorbing or masking the inhibition from the higher centers. The two left and right halves of the lumbosacral segment are engaged in reciprocal inhibition by virtue of increased activity following a mid- longitudinal cut. The increase in activity may be due to an inherent prop- erty of the cells after being released from inhibition, or it may simply be the result of an increase in the responsiveness of neurons in each half of the cord to the external environment. In either case, it is argued that bl; us the presence of cross fibers between the two half-segments is implied in the attenuation of activity. Similarly, one of the curious and unexpected findings is the increase in activity which accompanies deafferentation, an observation which sug- gests that peripheral feedback acts to damp spontaneous and long-term mo- tor output. It is not possible to predict from.this study where in the sensory system.this inhibition originates--- whether at the level of the sensory cells themselves or at the level of the receptor sites. The form- er would indicate inhibition through spontaneous activity of the sensory cells, while the latter would suggest responses to changes in the extern- al environment. From these findings it is concluded that not only are the inhibitory effects from the different parts of the nervous system.additive, but also that if it were possible to get an accurate measurement of the relative magnitude of’inhibition, it is suggested that the inhibition which origin- ates from.within a segment produces greater effects than that which is projected extra-segmentally. Pattern of‘Movements. Whereas the brain has been shown to have an influence on the activ- ity of the legs, it has virtually no effect on the pattern of movements. An integral lumbosacral cord is all that is required for the performance of alternate movements. than the integrity of the lumbosacral cord is in- terfered with, either through a mid-saggital section or by deafferentation, the alternate pattern is altered to that of simultaneous kicking of the legs . The one conclusion drawn from these observations is that sensory cells including their central terminals and, naturally, cross fibers 146 between the left and right half-segments are very essential for alterna- tion of legs to occur, much as they appear to be at the base of crossed extension reflexes. The simultaneous pattern is mediated by the lumbosacral segment alone since the exclusion of brain influence (MTTS + ISMSS and MTTS + ISDF) does not in any way alter the pattern. Because deafferented embryos ex- hibit the same (simultaneous) pattern as embryos with longitudinal cuts, it is tempting to look for a common mechanism, or common feature within a mechanism, through which this form of coordination is achieved. Yet in reality, the same pattern may be subserved by two entirely different mech- anisms. Since coordination occurs in the absence of sensory feedback, the information to propel the legs synchronously must consist of a central command. The command, moreover, makes use of the endsting cross fibers to set the rhythm of the moving legs. However, in embryos with complete mid- longitudinal cuts, the sensory system is still connected separately to each side. The information which determines the synchrony, it is suggest- ed, is one of mechanical factors by which afferents on each side are stim- ulated simultaneously. The position of the embryo within the remaining portion of the shell is such that the toes and tarsal Joints are braced against the inside of the shell, thus creating a mechanical friction. The resulting friction, then, comprises the simultaneous stimulus. If this should be the mechanism which operates in the mid-saggitally-seetioned em- bryos, then one can predict that prior to hatching, complete peeling of the shell and membranes, together with allowing the legs no access to mechanical surface, would result in asynchronous coordination. Similarly, one predicts that non-coordination will result from subjecting an embryo to mid-saggital section plus deafferentation. Consequently, the difference 117 between simultaneous pattern of coordination in deafferented and mid-sag- gitalJy-sectioned embryos may be that central factors are prepotent in the former, while the latter is subserved by peripheral factors. As was pointed out in the discussion section, another mechanism which might be operative in the deafferented embryos is the suggestion that peripheral feedback is not essential for the alternate pattern of coordination. The simultaneous pattern observed after deafferentation is simply the consequence of the central lesion, whereby shaving of the dor- sal portion of the lumbosacral cord removes, along with the target prim- ordial sensory cells, a pool of other cells, the normal function of which is to facilitate inhibitory coupling between motor centers on the two sides of the cord. One way to resolve the question concerning these two possible mechanisms is to develop a method of deafferentation which re- moves only the dorsal roots, while leaving the cord essentially intact. It is not possible at this Juncture to reconstruct events in devel- opment which lead to the commencement of coordination. It cannot be ascer- tained whether the final features are either central or peripheral or both, but this study reiterates the notion that the beginning of coordin- ated alternate movements does not originate simply in the establishment of functional connections between the two half-segments, but that other factors distributed in the entire lumbosacral cord are involved also. Given this supposition, the simultaneous pattern of coordination can be viewed as having undergone a development of itself, rather than merely a consequence of neural disruption. To put it in another way, it is quite conceivable that different kinds of results, in terms of rates of activ- ity or even pattern, may be obtained if the same lesions (i.e. ISMSS or LSIF) were performed in chick embryos just prior to hatching. Differences 148 in results may arise from the fact that in the developed chick, most fea- tures in the central nervous system are alreacbr "set" in and neural plas- ticity is all the more reduced. Despite the possible existence of a multitude of factors, it is very tempting to suggest that the final feature which chronicles the al- ternation of legs in normal embryos is the establishment of functional connections between sensory terminals and central neurons. This idea gains support from the established fact that in most developing neural systems in vertebrates, sensory mechanisms become funotional long after motor sys- tems have been in operation. GENERAL SUI-MARY Since it has been established that general as well as specific in- tegrated movements in the chick embryo do not appear until shortly after day 17 of incubation, it was proposed that the developnent of coordina- tion of leg movements could still be shom during a two-day interval, be- tween day 19 and hatching. Complementary to this inquiry, were the vari- ous manipulations of the central nervous system, which sought to provide an understanding of the neural mechanisms subserving coordination. The results from these two areas of investigation were deemed necessary be- fore strong statements or predictions could be made about the type of features (structural and/or functional) vhich alter to mediate develop- ment of coordination. During the two-day period, the degree of limb coordination (measured in terms of a ratio between coordinated movements and total activity) did not appear to change. However, a trend toward an increase in the rate of activity was observed in two experimental groups (figures I; and 5). The increase was attributed to a lesion connnon to both groups. No other at- tern of development relevant to limb coordination could be sham from day 19 onwards. Animals with mid-thoracic gaps, which serve to exclude the influ- ence of the brain from reaching the lumbosacral cord, are able to per- form alternate coordinated movements rhich are quite indistinguishable from those of the controls. However, the frequency of the movements of these spinal animals is significantly lower than that of the control group. ’49 SO Enbryes in which the lumbosacral segnent has been sectioned und- saggitally showed an increased rate of activity over that of the controls. It was concluded that the two halves of the cord segment reciprocate in inhibiting endogenous activity. The pattern of movements displayed by this group is that of the simultaneous kicking of the legs. Deafferentation of the lumbosacral area results in increased activ- ity over that of the controls, suggesting that the normal action of sens- ory cells is to inhibit endogenous motor output. An alternative interpre- _ um... nn‘ .471 tation is to propose that the form of deafferentation used in this stuw has the effect of removing cells (interneurons) which normally facilitate the inhibitory coupling between motor centers on the two sides of the I‘ll '- cord. The pattern of movements in the deafferented embryos is of the simul- taneous type. It is preliminarily concluded that sensory cells are neces- sary in mediating the alternate pattern of coordinated movements. Enbryos with both transverse sections in the mid-thoracic level and the mid-saggital sections of the lumbosacml cord perform coordinated movements at a much higher rate than the controls. Similarly, embryos with both mid-thoracic sections and deafferented lumbosacrals perform coordin- ated movements at a rate much higher than that of the controls. Since the rate of leg activity in the double lesioned animals is also greater than that of the corresponding single lesion (i.e. M'ITS + LSl-iSS ISI‘SS; ms + ISDF LSDF) , it was concluded that the brain acts to inhibit the lumbosacral cord; but that this inhibition is best detected after the dis- ruption of the lumbosacral cord. Furthermore, it was also concluded that segmental (lumbosacral) and extrasegmental (brain) influences were addi- tive. Like animals with longitudinal sections, or deafferented lumbosacrals, the double lesioned animals exhibit the simultaneous pattern of leg 51 mvements. Mechanisms which subserve the simultaneous pattern are suggest- ed. BIBLIOGRAPHY BIBIIOGRAPHI Abbott, U.K., and £1.14. Craig. 1963. The laboratory preparation of nor- mal avian embryos. Poultry Sci. , 1:2: h29-h37. Balaban, M., and J. Hill. 1969. Perihatching behaviors of chick em- bryos (Callus domesticus). Anim. Beh., 17: 350—359. Ballintijn, C.M. 1961. Experimental production of very small lesions by electrocoagulation. Experientia, 17: 1412. Bethe, A. 1930. Studien fiber die Plastizitat des Nervensystems. I. Mitteilung. Arachnoideen und Crustacean. Pfliig. Arch ges. Physiol., 22h: 793-820. Carmichael, L. 1926. The development of behavior in vertebrates exper- imentally renoved from the influence of external stimula- tion. PsyChe Revs, 33: 51-580 . 1927. A further start of the development of behavior in vertebrates experimentally removed from the influence of external stimulation. Psych. Rev. , 3h: Biz-147. Coghill, G.E. 1929. Anatomy and the Problem of Behavior. Cambridge University Press, London. Decker, J.D. 1970. The influence of early extirpation of the otocysts on development of behavior of the chick. J. Exp. 2001., 17,4: 3h9‘3a‘e Decker, J.D., and V. Hamburger. 1967. The influence of different brain regions on periodic motility of the chick embryo. J. Exp. 20010, 165: 371-381.}. Delcomn, F. 1971. Effect of limb amputation on locomotion in the cock- roach, Periplaneta americana. J. Exp. 3101., 5h:h53-h72. Detwiler, 3.12. 1920. Experiments on the transplantation of limbs in Amblystoma. Further observations on peripheral nerve con- nectiOnSO Jo RP. 20010, 35: 15. . 1936. Neuroembryology. McMillan Co., New York., pp.Sh-70. Doty, R.W., and J .F. Bosma. 1956. An electromyographic analysis of re- flex deglutition. J. Neurophysiol., 19: 1415-60. 52 53 Engberg, I., and A. Iundberg. 1962. An electromyographic analysis of stepping in the cat. Experientia (Basel), 18: 1714-177. Evdng, A.W., and A. Manning. 1966. Some aspects of the efferent con- trol of walking in three cockroach species. J. Insect Phys- iol., 12: 1115-1118. Fraser-Howell, C.H. 1963a. The relation between stimulus parameters and current flow through stimulating electrodes. J. Exp. Biale, 140: 15-220 . 1963b. A method for chronically implanting stimu- lating electrodes into the brains of locusts, and some re- sults of stimulation. J. Exp. Biol., no: 271 -28h. Gottlieb, G. 1963. Refrigerating eggs prior to incubation as a way of reducing error in calculating developmental age in imprint- ing experiments. Anim. Beh., 11: 290-292. Gray, J. 1950. The role of peripheral sense organs during locomotion in the vertebrates. Symp. Soc. Exp. Biol., 1:: 112-126. Gray, J ., and H.W. Lissmann. 19h6a. Further observations on the effect of deafferentation on the locomotory activity of amphibian MbSe J. Exp. B1010, 23: 121-1320 . 19h6b. The coordination of limb movements in am- phibia. J. Exp. Biol., 23: h15-h28. Hamburger, V. 1960. A Manual of Experimental Embryology. The Universi- ty of Chicago Press, Chicago. . 1963. Some aspects of the embryology of behavior. Quart. RBVO B1010, 38: 31.12-3650 . 1968. Emergence of nervous coordination. Origins of in- tegrated behavior. 27th Symposimn of Society for Develop- mental Biology. Dev. Biol. Supp. , 21: 251-271. . 1972. (In Press). Hamburger, V. , and H.L. Hamilton. 1951. A series of normal sta es in the development of the chick embryo. J. Morph., 8 : 149-92. Hamburger, V., and M. Balaban. 1963. Observations and experiments on spontaneous rhythmical behavior in the chick embryo. Devel. Biol., 7: 533-516. Hamburger, V., M. Balaban, R.W. Oppenheim, and E. Wenger. 1965. Period- ic motility of normal and spinal chick embryos between 8 and 17 days of incubation. J. Exp. Zool., 159: 1-1h. Sh Hamburger, V. , E. Wenger, and H.111. Oppenheim. 1966. Motility in the chick embryo in the absence of sensory input. J. Exp. Zool., Hamburger, V., and RM. Oppenheim. 1967. Pre-hatching motility and hatching behavior in the chick. J. Exp. Zool., 166: 171- 201:. von Holst, E. 1939. Entwurf eines Systems der Lokomotorischen Perioden- bildungen bei Fischen. Z. verg. Physiol., 26: h81-528. von Holst, E., and U. Saint Paul. 1962. Electrically controlled behav- iors 8010 Amara, 206: 50-590 . 1963. On flmctional organization of drives. Anim. Beh., 11: 1-20., translated from Naturwissenschaften, 18: 1109‘14220 Horridge, G.A. 1961. The centrally determined sequence of impulses in- itiateg from a ganglion of the clam M13. J. Physiol., 155: 320-33 . Hoy, 12.12., and D.M. Wilson. 1969. Rhythmic motor output in the leg mo- tor neurons of the milkweed bug, Oncopeltus. Fedn. Proc., Huber, F. 1959. Auslosung von Bewegungsmustern durch elektrische Reiz- ung des Oberschlundganglion bei Orthopteren. (Saltatoria: Gryllidae, Acridiidae). Verh. Zool. Ges. Munster, 2001. Am. Suppl., 23: 2,48-2690 . 1960. Untersuchungen 'uber die mnktion des Zentralnervensys- tems und inbesondere des Gehirnes bei der Fortbewegrmg und der Lauterzeugung der Grillen. Zeits. Vergl. Physiol. , ’41]: 60-1320 . 1962. Central nervous control of sound production and some speculations on its evolution. Evolution, 16: h29-hh2. . 1965. Neural integration (Central Nervous System) in lh__e_ W of Insecta, Rockstein, M. , ed. , Academic Press, 0 e, ppe 333-5%. Hughes, A., and M.C. Prestige. 1967. Development of behavior in the hindlimb of Xenopus laevis. J. Zool., London, 152: 3b?- 359. Hughes, 0.24. 1952. The coordination of insect movements. I. The walking movements of insects. J. Exp. Biol., 29: 267-281;. . 1957. The coordination of insect movements. II. The eff- ect of limb amputation and the cutting of commissures in the cockroach (Blatta orientalis). J. Exp. Biol., 31:006-333. SS Ikeda, K., and C.A.G. Wiersma. 1961:. Autogenic rhythmicity in the abdominal ganglia of the crayfish: control of swimneret movements. Comp. Biochem. Physiol., 12: 107-115. Kuo, Z.Y. 1967. Dynamics of Behavior Development: An Epigenetic View. Random House, New York. Lassek, A.M., and E.K. Mayer. 1953. An ontogenetic stucw of motor defi- cits following brachial rhizotomy. J. Neurophysiol., 16: 210-251. New, D.A.T. 1957. A critical period for the turning of hen's eggs. J. Embryol. Exper. Morph., 5: 293-299. Oppenheim, R.W. 1970. Some aspects of embryonic behavior in the duck (Anas Qatyrhynchos). Anim. Beh., 18: 335-352. Oppenheim, 12.14., and C.H. Narayanan. 1968. Experimental studies on hatching behaviors in the chick. I. Thoracic spinal gaps. Jo Expo 2001., 168: 395-11020 Oppenheim, R.W., R.J. Jones, and G. Gottlieb. 1970. Enhryonic motility and post-hatching perception in birds after prenatal gamma irradiation. J. Comp. Physiol. Psych., 71: 6-21. Pearson, K.G. 1972. Central programming and reflex control of walking in the cockroach. J. Exp. Biol., 56: 173-193. Pearson, K.G., and J.F. Iles. 1970. Discharge patterns of coxal lava- tator and depressor motor neurons of the cockroach, Peri- planeta americana. J. Exp. Biol., 51:: 215-232. Piatt, J. 1956. Studies on the problem of nerve pattern. I. Trans- plantation of the forelimb primordium to ectopic sites in Ambystoma. J. Exp. Zool., 131: 173-1914. Sherrington, 0.3. 1906. Integrative Action of the Nervous System. Cambridge University Press, London. . 1910. Flexion-reflex of the limb, crossed extension rgflex, and reflex stepping and standing. J. Physiol., no: 2 -121. . 1931 . Quantitative management of contraction in lowest 16%]. coordimtione Brain, 5h: 1-280 Siegel, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, New York. Sokal, R.R., and F.J. Rohlf. 1969. Biometry. The principles and prac- tice of statistics in biological research. W.H. Freeman and Co., San Francisco. 56 Strazrdcky, K. 1963. Function of heterotopic spinal cord segments in- vestigated in the chick. Acta. Biol. Hung., 11:: 1145-1 55. Straznicky, K., and G. Szekely. 1967. Functional adaptation of thor- acic spinal cord segnents in the newt. Acta. Biol. Acad. Sci. Hung., 18: 14119. Szekely, G. 1963. Functional specificity of spinal cord segments in the control of movements. J. Dubryol. Exp. Morph., 11: 167- 183. . 1967. Logical network for controlling limb movements in Urodela. Acta. Physiol. Acad. Sci. Hung. , 27: 285-289. Szekely, G., G. Czeh, and G. Voros. 1969. The activity pattern of limb muscles in freely moving, normal and deafferented newts. Exp. Brain Res., 9: 53-62. Szlep, G. 1952. On the plasticity of instinct of a garden spider (_A_ra- nea diadema) construction of a cobweb. Acta. Biol. Exp., T6? g‘2§e Tomita, A. 1967. A study on the movement pattern of four limbs in walking. II. EMG study on coordination of muscular act- ivities of the muscles of the four limbs during walk in man and dog. Zinmigaku Zassi, 75: 173-1914. In Japan- ese cited from Biol. Abst. 6950 (1968). Usherwood, P.N.R., and H.I. Bunion. 1970. Analysis of mechanical re- sponses of metathoracic extensor tibiae muscles of free- walking 1°m3te Jo Expo B1010, 52: 39-580 Usherwood, P.N.R., H.I. Runion, and J.I. Campbell. 1968. Structure and physiolog of a chordonotal organ in the locust leg. J. Expo 8101., 1.18: 305.323. Vowles, D.M. 1961. Neural mechanisms in insect behavior. in Current Problems in Animal Behavior, Thorpe, W.H. and 0. . Zang- me., Cambridge University Press, London. pp. 5-29. Weiant, E.A. 1956. Control of spontaneous efferent activity in certain efferent nerve fibers from the metathoracic ganglion of the cockroach. in Proceedings of the X International Entomo- logical Cong%ss Dome, ATER. and 111.5. Friend, eds. V0 e ’ ppe " 20 ’ Weiss, P. 19m. Does sensory control play a constructive role in the deveéopment of motor coordination? Schweiz. Med. Wochenschr., 71 : 91 '5950 . 191:1 b. Self-differentiation of the basic patterns of coordin- ation. Comp. Psychol. Monogr., 17. 57 ‘Weiss, P. 1950. Experimental analysis of coordination by the disarr— angement of central-peripheral relations. Symp. Soc. Exp. Biol., h: 92-111. . 1950. The deplantation of fragments of nervous system in am- phibia. I. Central reorganization and formation of nerves. J. Exp. Zool., 113: 397-b61. Wendler, G. 1966. The coordination of walking movements in anthropods. Symp. Soc. Dcp. Biol. No. 20. Nervous and hormonal mech- anisms of integration. pp. 229-250. Wenger, B. 1968. Construction and use of the vibrating needle for em- bryonic operations. Bio. Science, 18: 226-228. g H Wilson, D.M. 1961. The central nervous control of flight in a locust. Jo Expo B1010, 38: h71"h90e Wilson, D. , and E. Gettrup. 1963. A stretch reflex controlling wing beat grequency in the grasshopper. J. Exp. Biol. , ho: 171‘1 So APPENDIX This appendix consists of a compilation of raw data from.which the various values presented in the different experiments were derived. These raw data have been arranged to follow, more or less, the same or- der in which the transformed data were presented in the body of the the- sis. That is to say, there are three sets of data: the one set deals with the temporal pattern of activity in terms of absolute rates; anoth- er set concerns the temporal pattern of activity with respect to the magnitude of limb coordination, measured as a ratio (percentage) between coordinated limb movements and total leg activity; and the last set con- sists of activity peak values, out of which the various groups of embryos were compared. Except for the last set of data (Tables 13-18), the suc- cessive values in the tables represent the temporal performance of each embryo at that particular stage of development. 58 59 Tables 3-7 contain the raw data for the frequency of leg activity as a function of developmental time. The actual numbers represent the number of coordinated movements per minute. As with all tables, the legend for each group is provided at the bottom of the table. Ell» 13...! . Le mw.m mo.~ mm.m U\b~b~ O O 000% MN Ale—.3 MMON O\(\ [\N—IT com .qmo (I) .3 m 0 22m .83 mum 8% men an 49m 8; mw.m w~.o e e e J—ij—Jv—JOP mOPL‘DQD F-WO r-b-JOOOUAe-N O\-:T O NN.F 8.4 2.0 mm.F one sum mm.m uo.w ow.N mad 2; mm.e mmé mus Pm om we NF om \O 1— ZOHB040-o\U\ux03o\o\ :..... ......... OFF-1"!— r-Nq-ONNONN mammm CDO\ m0mwm \o ..... .. ..... v-r-r-v-v- PO FNNV‘O gym com (\I \01- O m .. .. . .. 0 CO CO 1- O‘- N O . 1— ..:t P 0 :3 W: G! InO—ZTN (DmMMON O\(”\O\ E-« F) 000E0§000080 Ox 5 NNI—v-f- Mid-thoracic transverse section. TIME IN HOURS FROM PEAK Table 1 S. 714 CHICK # O ‘— \O N Chm F0 «3 0 . 0 0 . 0 O O .301 N‘— 00\ .3 O\ NO\O q—N \o . 0 0 . . 0 0 0 0 c—O .— I- JEN 00.: r-O\(-\O\O.:I MIA \OOLn hm _:f . . . 0 . 0 0 . . 0 . 0 0 NONFPN JN Jmm 0\—:T OFMNO—UOJOr—ONMONJQ N . 0 . . 0 0 0 . . 0 . 0 0 . 0 0 0 NNMrMMNQNmOmQ—flr—mc— :— «— N010 QJMNNb-OQaDr-c—(DCDCD m . . . 0 0 . 0 0 0 . . 0 0 0 0 0 . NMM—dmmWWONNQOPFF-M v-s-I- ‘— N mxomoozoommxoaowmmeomoo NNNPONOMMJJmMNJMQ O b—Ovo N NM!“ qu- O\ J 0 0 0 . 0 . . . 0 0 0 N Nq—o m mmm (“‘0 N O \O .301 0.3 \o 0 . . . 0 0 N m :14“ NM m max O\O\ w . 0 0 0 . O N0- N1- 0 C- O b—Nm M OxOCDONb-b-w {Rm-SMNQO'SORF-v-NQD \omm J FN Nm NmNN J3!“ Lumbosacral mid-saggital section. TIME IN HOURS FROM PEAK Table 16. 7S CHICK# O o 0 r- ‘— l~ \o 00 m 0 . 0 O O b- CI) Ox CO b-O\M O\ \o . 0 . 0 0 . . O 1— !- v-m\0 In C (“P—3 b-NmCDN—ZTN J . 0 0 0 0 0 . . . 0 0 N NNq- mqmnmxoxo N r-O\O\MN\O[\CDc—O.:.TL\NI\- MNMNMM—é—éIn’l-nInF-(Do In\OOv—In.:f\OO\CDInCD\OO\‘- ‘1‘ 0 0 . 0 . 0 0 0 . 0 0 0 . 0 mmzqzmmmxowb-mmb- ,— N O\\OO\COO\CO-Inl\N (\I ...0..0000.0.00. NMMInJMMFNNJF-OJ—dv- FFF‘I— Q%WQ%%%QQW%QQfifiW £14 mm\o\0\0\0t~oocoooooo\omo\o\ I—C—r'r- 9Q€W€Wfi€fifi$99%9@ N 41.: Mina—3.3.30 —:T