An autoradiographic study of the chick brain after imprinting

Brain Research, 168 (1979) 361-373 © Elsevier/North-Holland Biomedical Press

361

AN AUTORADIOGRAPHIC STUDY OF THE CHICK BRAIN AFTER IMPRINTING

G. HORN, B. J. McCABE and P. P. (3. BATESON

University of Cambridge, Department of Zoology, and Sub-Department of Animal Behaviour, Cambridge ( U.K.) (Accepted September 14th, 1978)

SUMMARY

On the first day after hatching domestic chicks were exposed to an imprinting stimulus, a horizontal yellow slit of light moving upwards in a window and presented at a rate of 4 slits/sec. Chicks were exposed for either 45 rain (undertrained) or 180 rain (overtrained) on the first day of hatching (60 or 240 min in the case of 1 pair). On the second day all birds were exposed for a further 63 rain. Twenty birds were matched in pairs and each chick received 1.1/zCi [14C]uracil/g body weight injected into the heart region before exposure on day 2. At 150 rain after the injection the chicks were decapitated and serial coronal sections of their brains cut; alternate pairs of sections were prepared for autoradiography. The optical density for a number of major anatomical regions was measured. The measurements for each region were averaged over _~ 0.6 mm 'slabs' of brain and expressed as a percentage of the mean of all measurements for that brain. Standardized mean optical density was significantly greater in undertrained chicks than in overtrained chicks in a part of the medial region of hyperstriatum ventrale (MHV) which extended across two adjacent slabs. The slabs were slightly posterior to the mid-point between anterior and posterior poles of the brain. In the next two, adjacent, anterior slabs the variances for MHV were significantly greater in undertrained chicks than in overtrained chicks. There were no other significant differences between brain regions. Taken together with previous studies these results suggest that the intermediate and medial part of hyperstriatum ventrale is intimately linked with the imprinting process.

INTRODUCTION

The recently hatched chick will quickly restrict its social preferences to a conspicuous object as a result of being exposed to it. The learning process is called

362 ,imprinting, l,32. After young chicks have been exposed to a flashing, rotating light, which is a highly effective imprinting stimulus 9-11, a variety of biochemical changes occur in a particular region of the brain~,~7, '~9. These changes might be related specifically to information storage since the chick learned the characteristics of the stimulus; but they may also reflect non-specific changes in the CNS resulting, for example, from differences in movement, stress and sensory stimulation between the trained and untrained birds. A series of experiments have suggested, however, that the net incorporation of radioactive uracil into acid-insoluble substances in the 'roof' of the forebrain is intimately associated with the learning process rather than with other consequences of the imprinting procedure6,S, 19,2°. The results were sufficiently encouraging to justify a detailed analysis of the morphological and physiological changes which the gross biochemical events reflect. Such an analysis requires precise information concerning the site of the biochemical changes. The forebrain roof is a relatively large mass of tissue and includes a number of anatomically distinct regions. Are training-dependent biochemical changes uniformly distributed throughout this region or are they localized? It is necessary to distinguish between these possibilities before proceeding to a more detailed level of analysis. The purpose of the present autoradiographic study was to specify the sites of biochemical changes resulting from imprinting, if indeed such changes are localized within the forebrain roof of the chick's brain. The design of the present study was based upon that of an earlier experiment s. In that study, groups of chicks were trained for different periods on the first day after hatching (day 1); it was found that the longer the period of training the stronger was the preference for the familiar stimulus. The chicks were then exposed for 1 h to the imprinting stimulus on the second day after hatching (day 2). We presumed that those chicks exposed to the stimulus for the longer period on day 1 would have less to learn about the stimulus than chicks exposed for the shorter period on day 1. We found on day 2 a higher incorporation of radioactive uracil in the forebrain roof of the birds trained for the shorter period on day 1. This experimental design enables chicks to be selected and matched in pairs on the basis of their behavioural activities. Since autoradiographic analysis is time-consuming we needed a method for discarding behaviourally inactive chicks. However, the same method of selection had to be applied to both groups whose brains were to be compared in order to avoid certain difficulties in interpreting any differences between them. For example, if the brains of behaviourally active imprinted chicks were compared with those from dark-reared chicks selected at random - - and therefore including inactive b i r d s - differences between the brains of the two groups could arise from the statistically greater behavioural activity of the imprinted birds. When the behavioural variation is reduced by selecting and matching chicks in pairs in this way, the statistical significance of the biochemical differences between the two groups is greatly increased 7. A brief account of the present study has been published z.

363 MATERIALS AND METHODS

Subjects Four batches of domestic chicks (Ross Chunkies) were used. The methods for incubating, hatching and handling have been described elsewhereL After selection (see below) brains from 20 chicks, matched in pairs, were used for autoradiographic analysis. All chicks were separately exposed to a static overhead white light for 30 rain prior to training la. Training with a moving pattern began 15-25 h after hatching. The apparatus for training the chicks is described in detail elsewhere 2. Each chick was placed in a running wheel situated 40 cm away from the training stimulus. This consisted of a horizontal yellow slit (2 × 15 cm) moving upwards in a window 10 cm high at a maximum rate of 47 cm/sec. The stimulus presented 4 upward moving slits per see. Above the flashing light was a loudspeaker which emitted the maternal call of the domestic hen at an intensity of 65 dBA measured from the chick's position. Training was conducted in a cabinet maintained at 29.8-30.2 °C and lit by an overhead bulb (25 W). The flashing light, the maternal call and the overhead bulb were all controlled by the same timer. Each chick's approach and withdrawal activities were separately recorded. With the exception of one batch all chicks were initially put into the wheels where they received 45 min of exposure to the training stimulus and after training they were returned to a dark incubator kept at 33-34 °C. Half of each batch of chicks received only one 45 rain period of training on the first day after hatching: these are called 'undertrained'. The other half, called 'overtrained', were returned to the wheels until they had received a total of 180 rain of exposure to the training stimulus. One batch (from which one matched pair of brains was subsequently obtained) were treated slightly differently on the first day after hatching. The flashing light was not accompanied by the maternal call and the periods of exposure were longer. The undertrained chick was exposed for 60 min and the overtrained for 240 min. All birds were treated identically on the second day (37-48 h after hatching). Each chick was weighed and injected in the heart region with [2-14C]uracil (59 mCi/mmol) in 154 mmol saline. A 45 g bird received 50/~Ci of radioactive uracil in 0.2 ml saline and other birds received a dose in direct proportion to their weight. After injection each bird was placed in a dark incubator maintained at 29.5-30.5 °C; 47 rain later it was placed in a training wheel and exposed to the training stimulus for 63 min. It was then withdrawn from the wheel and returned to the dark incubator kept at 29.5-30.5 °C. Finally, 150 rain after injection the bird was decapitated, the whole brain was dissected out and frozen in solid CO2. The brain was then placed into a coded jar containing a small quantity of frozen water to retard subsequent dessication of the brain and stored on solid COs. Only chicks that were active on the first day - - about half of the total - - were picked for retraining on the second day. Moreover, chicks that failed to reach a criterion of averaging 10 approach counts/min on the second day (approximately equivalent to travelling at a rate of 10 m/rain) were discarded. Previous studies had shown that incorporation of radioactively labelled uracil into macromolecules of all regions was strongly dependent on the approach activity of

364 the chicks 6. In order to minimize such effects, which are superimposed on those of the training procedure, undertrained and overtrained chicks were, as far as possible, matched in pairs on the basis of their approach activity during training (see Table !). Furthermore, since dilution of the radioactive uracil in the blood might not be directly proportional to weight, undertrained and overtrained birds were matched in terms of their weight on the second day after hatching (see Table I). All selection and matching was done before autoradiographic analysis was started and those people involved in the autoradiography were unaware of the history of each brain.

Autoradiography Each matched pair of brains was placed in a cryostat at --30 °C. Serial coronal sections 15 /~m thick were cut from each brain and mounted on glass slides, two consecutive sections per slide. Alternate pairs of sections from each brain were fixed in 10 % formol saline and stained with cresyl violet: the remaining sections were prepared for autoradiography 18. In 5 of the 10 pairs of brains the brain from the undertrained chick was cut before the brain from the overtrained chick. This order of cutting was reversed for the remaining 5 pairs of brains. The order of cutting was unknown to the person who analyzed the autoradiographs. After drying, the most anterior 9 pairs of sections from each brain were placed against a sheet of Kodirex X-ray film (Kodak). The most anterior pair of sections from the brain of one chick was placed next to the most anterior pair from the other chick, and so on. All other sections from the two brains were placed alternately against sheets of film in this manner. The film and glass slides were clamped between two flat glass plates and stored in a lightproof box at room temperature. All sections from any one pair of brains were exposed for the same time. The mean exposure time was 13 weeks. Non-radioactive sections were placed against unexposed film and experimental radioactive sections placed against deliberately fogged film. These controls showed that positive chemography, negative chemography and pressure effects were absent. Development of the autoradiographs was conventional. The optical density of the autoradiographs was measured in order to provide an index of the radioactivity in the sections 16. A Joyce Loebl double beam microdensitometer was used for this purpose, employing a square light spot of 0.3 mm side on the film. Parts of each autoradiograph were easily identified as corresponding to different TABLE I

Approach counts and weights of 10 overtrained and 10 undertrained chicks Means and standard errors are shown. The chick pairs were matched on the basis of weight and behavioural activity during training. The matched pairs t values for approach activity on days 1 and 2 and for weight are given. None of these values is statistically significant.

Undertrained Overtrained Matched pairs t

Day 1 Day 2 Approach counts/min Approachcounts/min

Weight (g)

9.66 ± 1.11 9.11 ± 1.13 0.645

39.50 d= 1.49 40.05 ± 1.18 0.892

15.39 ± 1.28 16.12 ± 1.19 0.478

365 anatomical regions (cf. Fig. 3) and the optical densities of selected anatomical regions of the forebrain (see below) were measured. The optical density of film adjacent to each autoradiograph ('background') was substracted from each reading. The optical densities of all measured parts of each autoradiograph were less than 1.0. Each anatomical region selected for measurement was subdivided into one or more coronal 'slabs', approximately 0.6 m m thick and corresponding to 18 section autoradiographs from a brain. Within each slab measurements on 10 section autoradiographs were averaged and the mean value was then standardized by expressing it as a percentage of the mean of all measurements for that brain. For each region, the number of slabs was the same for all brains. The locations of sampling sites are shown in Fig. 1. Selection of sites was done on theoretical grounds and on the basis of a pilot experiment. In this experiment 6 pairs of chicks were trained on a schedule similar to that of the chicks in the main experiment. The section autoradio-

HA 1 HI/HD

B

~

M

H

~

V

1

~--

BASE 1

~~

5

~

~"

~

--BASE5

POST

t

5mm

1

NEO I

I

Fig. 1. A-D : tracings of section autoradiographs showing the position of the sampling sites in the most anterior slab of each anatomical region sampled. HA, hyperstriatum accessorium; HI/HD, hyperstriatum intercalatus/hyperstriatum dorsal¢; MHV, medial hyperstfiatum ventrale, the sampling point is dorso-medial to lamina hyperstriatica; NEO, neostriatum; ECTO, ¢ctostdatum; ARCH, archistriatum; POST. NEO. posterior neostriatum. The numbers refer to the anteroposterior position of the slab (cf. Table II).

366 TABLE I[ Summary o f results for standardized data The 8 anatomical regions were divided into 1 or more coronal slabs each approximately 0.6 mm thick. The slabs for any one region were adjacent to one another and are numbered with regard to their antero-posterior position: lower numbers denote more anterior positions. Standardized mean optical densities are given, averaged over all 20 chicks, together with the mean differences between overtrained and undertrained chicks matched in pairs and the standard errors of these mean differences. At each sampling site results from undertrained birds were compared with results from overtrained birds using a matched pairs t-test. Anatom~al region Slab Hyperstriatum accessorium Medial hyperstriatum ventrale

1 2 1 2 3 4 5 6 7 Neostriatum 1 2 3 4 5 6 7 Base 1 2 3 4 5 6 7 Hyperstriatum 1 intercalatus/dorsale 2 Ectostriatum 1 2 Post. neostriatum 1 Archistriatum I ** P -= 0.0056;

Standardized mean opt~al density (n -- 20)

Difference in standardized mean opt~al dens~y ( undertramed-overtrained, n -~ 10)

Mean

Mean

S.E.

89.53 88.48 124.44 121.98 118.58 116.51 118.32 117.60 113.75 107.21 104.73 99.63 96.17 94.06 91.53 88.57 90.09 81.74 71.83 65.76 61.92 60.54 63.12 112.91 109.36 146.76 150.00 89.37 105.39

--5.34 --5.44 0.26 1.14 0.09 2.64 6.23*** 6.27** --0.34 --2.08 -4).29 --2.26 --2.56 --0.89 --0.55 --0.77 --1.75 1.96 --1.45 0.04 --0.21 --0.02 --0.76 --1.93 3.03 3.85 3.13 --2.31 0.34

3.47 2.77 1.56 2.02 2.13 2.25 1.04 1.74 2.07 2.60 2.49 1.13 1.49 2.69 1.90 2.43 1.82 2.40 3.23 2.08 2.15 2.24 1.23 2.10 2.1 l 4.23 3.26 2.23 2.64

*** P = 0 . 0 0 0 2 .

graphs were studied by visual inspection and the training schedule, unknown to the observer, of each bird of a pair predicted. No decision could be reached for one pair of chicks. Of the remaining 5 pairs, the training schedules were correctly predicted in 4 of them. In all of these 4 pairs of chicks the medial part of the hyperstriatum ventrale (MHV), especially in the intermediate antero-posterior (A-P) region, was darker in the undertrained than in the overtrained chicks: no other difference between pairs of brains was as predictive of behaviour. The result of using MHV as predictor, though

367

*de

iii ¢O z LLI n-

u. O iii

1

2

3

4

SLAB

NUMBER

5

6

7

DIFFERENCE IN STANDARDIZED MEAN OPTICAL DENSITY (UNDERTRAINED-- OVERTRAINED) ACCORDING TO BRAIN SLAB (0.6mm) ~ #~

P< 0.01 P < 0.001

Fig. 2. Mean differences in standardized optical density at the sampling sites within MHV. The approximate locations of the sampling sites are shown on a diagram of a chick brain, the dorsal surface of which has been removed. Bars represent S.E.M.

Overt rained

Under t rained

Ventricle MHV

Neostriatum Archistriatum Lateral Hypothalamic area t

J

5mm

369 not of course statistically significant, was suggestive, so that in the main experiment M H V was sampled. In addition, the ectostriatum, hyperstriatum accessorium (HA) and hyperstriatum intercalatus/hyperstriatum dorsale ( H I / H D ) 38 were sampled since these are visual projection areasZl, 23,~5-2s. Measurements were made for neostriatum, part of which has been compared by Cohen and Karten 1~ with the intrinsic regions of the mammalian neocortex. The optical density for archistriatum was also measured. This complex was easily identified on the section autoradiographs, but it was not possible to distinguish the individual nuclei which together constitute the archistriatum 8~. Finally, on every section autoradiograph the optical density was measured for a region situated in the ventromedial part of a hemisphere well within the forebrain 'base'Z. Different nuclei were selected varying according to the position of the section autoradiograph in the A-P plane, but for a given position in this plane was the same for each brain of a pair. RESULTS Stadardized mean optical density was significantly greater in undertrained chicks than in overtrained chicks in a part of M H V which extended across two adjacent slabs (Table 1I). The approximate positions of these two slabs are shown in Fig. 2. It can be seen that the effect extends across an intermediate part of MHV, immediately posterior to the mid-point between the anterior and posterior poles of the forebrain. Sample autoradiographs from a pair of brains are shown in Fig. 3. The sections which produced these autoradiographs were taken from 'slab 6'. M H V was the only region in which statistically significant differences were found between the two groups of birds. However, when 29 comparisons are made a certain number of differences may appear, by chance, to be statistically significant. In slab 5 the difference between the mean optical density of the undertrained compared with the overtrained chicks was significant at P = 0.0002. Suppose that over 29 comparisons we wish to use an overall probability of P ---- 0.05 to give significance. We therefore accept as significant any single comparison for which P < 0.05/29 ----0.0017. The value of P ~ 0.0002 for slab 5 is clearly less than this and the difference therefore remains highly significant. Inspection of the data suggested that in the slabs of M H V in which no significant differences were found between undertrained and overtrained chicks, the variance was greater in the undertrained chicks. Accordingly, the variance of data from each group of

Fig. 3. Autoradiographs from overtrained and undertrained chicks of a pair. The diagram shows the parts of the autoradiographs sampled with the microdensitometer, The autoradiographs from the overtrained bird are slightly lighter overall than those from the undertrained bird. This effect, suggesting a lower incorporation of [14C]uracil into all areas of the overtrained chick, was not consistently distributed between chick pairs. The difference could be due to a variety of factors such as small differences in the amount of precursor injected. Variation due to such factors was eliminated by the standardization procedure used. The medial part of hyperstriatum ventrale is darker relative to surrounding regions in the autoradiographs from the undertrained chicks compared with those from the overtrained chicks.

370

I.k O O:

07

X

1

2

3

4

5

6

7

SLAB NUMBER VARIANCE RATIOS (UNDERTRAINED/OVERTRAINED) CALCULATED FOR EACH SLAB WITHIN M.H.V. "X- P< 0.05

Fig. 4. Variance ratios calculated from the standardized data for each of the 7 slabs within MHV,

Ordinate: varianceof the data from undertrainedchicks dividedby the variance of the data from the overtrained chicks. chicks was calculated for each slab within MHV and a variance ratio calculated by dividing the variance of data from undertrained chicks by the variance of data from overtrained chicks. The F values are plotted in Fig. 4. The variance is significantly greater in undertrained chicks in slabs 3 and 4. The variances were not significantly different in the remaining slabs within MHV, nor were there significant differences in the 22 sampling sites outside MHV. DISCUSSION

The optical density, and hence incorporation of [14C]uracil into acid-insoluble substances, was higher in a restricted region of the forebrain of undertrained chicks compared with overtrained chicks. This region is the medial part of hyperstriatum ventrale and extends at least 1 mm in the A-P plane lying just caudal to the mid-point between the anterior and posterior poles of the forebrain. Since MHV lies in the anterior part of the forebrain roof e,8 the present study confirms and extends the earlier studies. In turn, the results of those studies are applicable to the present work and suggest that the biochemical changes are closely related to the imprinting process. The autoradiographic technique used does not permit localization at a cellular level, for example into neurones and/or glial cells. Presumably the bulk of incorporation, a measure of turnover and not necessarily of synthesis, is into macromolecules in cell bodies. If these biochemical changes lead ultimately to changes in cell structure, the effects may be expressed in MHV neurones, the axons and dendrites of which ramify near to the cell body, and/or those whose processes pass more distally, ramifying into surrounding structures or even further afield. If the inferred cellular changes form the basis of information storage, this may occur during the imprinting process within MHV and/or in areas to which MHV projects. A recent study4 has shown that bilateral lesions to MHV severely impair imprinting.

371 In a recent paper al Salzen et al. report that ablation of the most anterior part of the forebrain roof, a lesion which included part of the hyperstriatum accessorium, had only a small effect on retention when chicks had been imprinted on a stationary coloured object. Neither in this nor in an earlier study by this group 30 was MHV destroyed. Salzen et al.30,31 found that retention was affected to a much greater degree with laterally placed lesions. However, following the lesion the chicks were very inactive and this may have affected their behaviour in the test. In addition, it is perhaps worth pointing out that Salzen et al. 30 state: ' . . . we cannot say whether the lateral lesions interfered with complex visual discrimination processes per se or with associative processes necessary for the formation of learned discriminations.' In some slabs in which no significant differences were found between under- and overtrained chicks in MHV, the variances were greater in the undertrained birds. This finding suggests that while changes are almost invariably produced in some parts of MHV by the imprinting procedure, other parts of MHV are only implicated in some individuals. The effect of individual variation in the extent to which certain regions of MHV are involved would have been to increase the variability in the undertrained birds in those regions. It is worth noting that the regions are adjacent to those slabs in which there are significant differences between mean values for over- and undertrained chicks. A possible interpretation of our findings is that the lower incorporation of uracil into the MHV of the overtrained birds was produced by greater amounts of auditory stimulation on the first day. We do not think this is plausible for at least two reasons. First, the birds were not isolated from sound when they were not being trained; they were able to hear the calls of other chicks and, of course, their own calls from before hatching until they were killed. This experience almost certainly swamped the difference between overtrained and undertrained birds in auditory experience associated with the imprinting procedure on the first day after hatching. Secondly, experiments of very similar design to the present one have clearly implicated the specific effects of visual experience on changes in uracil incorporation into the forebrain roof 6,8,2°. In those experiments, differences in brain biochemistry between chicks could not be attributed to differences in auditory experience. It is hardly likely that the present experiment uncovered a new phenomenon and simultaneously the previously discovered effects of visual experience evaporated. Evoked responses to visual stimuli14,2z or to electrical stimulation of the optic pathways2s have been recorded in the intermediate A-P region of hyperstriatum ventrale. In a recent studylz using horseradish peroxidase, it was shown that MHV receives connections from other telencephalic structures including HA, HI/HD and neostriatum. In addition there is a major source of fibres passing directly to MHV from the optic tectum. These results show that MHV receives visual inputs from forebrain and mid-brain structures and possibly auditory as well as somatosensory inputs from the neostriatum. It is of considerable interest therefore, that the strength and speed of imprinting to a visual stimulus is facilitated by simultaneously presenting an auditory stimulus (see ref. 1). MHV has recently been shown to project to the paleostriatal complex, to parts

372 of the archistriatum a n d to the lateral corticoid area 13. A l t h o u g h the paleostriatal complex has been c o m p a r e d with the basal ganglia of m a m m a l s 24, the role of the avian p a l e o s t r i a t u m in the control of m o v e m e n t is n o t clear. The archistriatum is considered by Zeier a n d K a r t e n 3~ to be subdivided into 'Iimbic' a n d 'somatic s e n s o r i m o t o r ' c o m p o n e n t s , the latter being c o m p a r e d b o t h by these authors a n d by Zecha 84 with the s e n s o r i m o t o r cortex of primates. A l t h o u g h the significance of the projection to the lateral corticoid area is unclear M H V appears to be r e m a r k a b l y well-placed to integrate a variety of sensory inputs a n d to exercise some control over m o t o r activity. ACKNOWLEDGEMENTS We are indebted to Professor K. Webster for valuable discussions a n d to the S.R.C. for financial support.

REFERENCES 1 Bateson, P. P. G., The characteristics and context of imprinting, Biol. Rev., 41 (1966) 177-220. 2 Bateson, P. P. G., Brief exposure to a novel stimulus during imprinting in chicks and its influence on subsequent preferences, Anim. Learn. Behav., (1979) in press. 3 Bateson, P. P. G., Horn, G. and McCabe, B. J., Imprinting and the incorporation of uracil in the chick brain: a radioautographic study, J. Physiol. (Lond.), 275 (1978) 70P. 4 Bateson, P. P. G., Horn, G. and McCabe, B. J., Imprinting: the effect of partial ablation of the medial hyperstriatum ventrate of the chick, J. Physiol. (Lond.), 285 (1978) 23P. 5 Bateson, P. P. G., Horn, G. and Rose, S. P. R., Effects of early experience on regional incorporation of precursors into RNA and protein in the chick brain, Brain Research, 39 (1972) 449-465. 6 Bateson, P. P. G., Horn, G. and Rose, S. P. R., Imprinting: correlations between behaviour and incorporation of [14C]uracilinto chick brain, Brain Research, 84 (1975) 207-220. 7 Bateson, P. P. G., Horn, G. and Rose, S. P. R., (1979) in preparation. 8 Bateson, P. P. G., Rose, S. P. R. and Horn, G., Imprinting: lasting effects on uracil incorporation into chick brain, Science, 181 (1973) 576-578. 9 Bateson, P. P. G. and Jaeckel, J. B., Imprinting: correlations between activities of chicks during training and testing, Anita. Behav., 22 (1974) 899-906. l0 Bateson, P. P. G. and Jaeckel, J. B., Chicks' preferences for familiar and novel conspicuous objects after different periods of exposure, Anim. Behav. 24 (1976) 386-390. 11 Bateson, P. P. G. and Wainwright, A. A. P., The effects of prior exposure to light on the imprinting process in domestic chicks, Behaviour, 42 (1972) 279-290. 12 Bradley, P. and Horn, G., Afferent connections of hyperstriatum ventrale in the chick brain, J. Physiol. (Lond.), 278 (1978) 46P. 13 Bradley, P. and Horn, G. Efferent connections of hyperstriatum ventrale in the chick, J. Anat. (Lond.), in press. 14 Brown, M. W. and Horn, G., Effects of visual experience on unit responses in hyperstriatum of chick brain, J. PhysioL (Lond.), 278 (1978) 48P. 15 Cohen, D. H. and Karten, H. J., The structural organization of avian brain: an overview. In D. H. Cohen and D. P. Pitts, (Eds.), Birds, Brain andBehaviour, Academic Press, New York, 1976, pp. 29-73. 16 Goldstein, D. J. and Williams, M. A., Quantitative autoradiography: an evaluation of visual grain counting, reflective microscopy, gross absorbance measurements and flying-spot microdensitometry, J. Microsc., 94 (1971) 215-239. 17 Haywood, J., Rose, S. P. R. and Bateson, P. P. G., Changes in chick brain RNA polymerase associated with an imprinting procedure, Brain Research, 92 (1975) 227-235. 18 Horn, G. and McCabe, B. J., An autoradiographic method for studying the incorporation of uracil into acid-insoluble compounds in the brain, J. Physiol (Lond.), 275 (1978) 2-3P. 19 Horn, G., Rose, S. P. R. and Bateson, P. P. G., Experience and plasticity in the central nervous system, Science, 181 (1973) 506-514.

373 20 Horn, G., Rose, S. P. R. and Bateson, P. P. G., Monocular imprinting and regional incorporation of tritiated uracil into the brains of intact and 'split-brain' chicks, Brain Research, 56 (1973) 227-237. 21 Hunt, S. P. and Webster, K. E., Thalamo-hyperstriate interrelations in the pigeon, Brain Research, 44 (1972) 647-651. 22 Jones, S. J. and Horn, G., Effects of visual experience on visually evoked potentials recorded in the chick forebrain, Brain Research, 159 (1978) 297-306. 23 Karten, H. J., The organisation of the avian telencephalon and some speculations on the phylogeny of the amniote telencephalon, Ann. N.Y. Acad. ScL, 167 (1969) 164-179. 24 Karten, H. J. and Dubbeldam, J. L., The organisation and projections of the paleostriatal complex in the pigeon (Columba livia), J. comp. Neurol., 148 (1973) 61-90. 25 Karten, H. J. and Hodos, W., Telencephalic projections of the nucleus rotundus in the pigeon, J. comp. Neurol., 140 (1970) 35-52. 26 Karten, H. J., Hodos, W., Nauta, W. J. H. and Revzin, A. M., Neural connections of the 'visual wulst' of the avian telencephalon. Experimental studies in the pigeon (Comlumba livia) and owl (Speotyto conicularia), J. comp. NeuroL, 150 (1973) 253-278. 27 Meier, R. E., Mihailovic, J. and Cu6nod, M., Thalamic organisation of the retino-thalamo-hyperstriatal pathway in the pigeon, Exp. Brain Res., 19 (1974) 351-364. 28 Revzin, A. M. and Karten, H., Rostral projection of the optic tectum and nucleus rotundus in the pigeon, Brain Research, 3 (1966/67) 264-276. 29 Rose, S. P. R., Early visual experience, learning and neurochemical plasticity in the rat and the chick, Phil. Trans. B., 278 (1977) 307-318. 30 Salzen, E. A., Parker, O. M. and Williamson, A. J., A forebrain lesion preventing imprinting in domestic chicks, Exp. Brain Res., 24 (1975) 145-157. 31 Salzen, E. A., Parker, D. M. and Williamson, A. J., Forebrain lesions and retention of imprinting in domestic chicks, Exp. Brain Res., 31 (1978) 107-116. 32 Sluckin, W. Imprinting and Early Learning, 2nd Edition, Methuen, London, 1972, 182 pp. 33 Tienhoven, A. van and Juhasz, L. P., The chicken telencephalon, diencephalon and mesencephalon in stereotaxic coordinates, J. comp. Neurol., 118 (1962) 185-197. 34 Zecha, A., The 'pyramidal tract' and other telencephalic efferents in birds, dcta morph, neerLscand., 5 (1962) 194-195. 35 Zeier, H. and Karten, H. J., The archistriatum of the pigeon: organisation of afferent and efferent connections, Brain Research, 31 ( 1971) 313-326.