Food and food consumption of nestling tits Food and food consumption of nestling tits, Parus major minor and Parus varius varius, in the ever-green broad leaved forests in northern Kyushu Kazuhiro Eguchi* Introduction This paper is concerned with the food consumption of nestling tits, Great Tit (Parus major minor) and Varied Tit (Parus varius varius), in the ever-green broad leaved forests. Measuring the food consumption of the nestling may provide an important cue to elucidate the range utilization of adult birds, but there have hitherto been few studies on this problem. In the case of song birds in the forest, methodological difficulties have prevented the study on feeding ecology in advance. Of song birds, Paridae often use artificial nest boxes for breeding. Taking advantage of this habit, food or food consumption of the nestling of Paridae, particularly great tit, has been assessed in various habitats by means of the direct observation or automatic photography (Betts, 1955, Tinbergen, 1960, Gibb & Betts, 1963, Royama, 1966). On the other hand, it is well-known that species belonging to Paridae coexist in a same forest segregating their foraging site vertically in trees (Hartley, 1953, Gibb, 1954, 1960, Nakamura, 1970, Ogasawara, 1970, 1974). Great tit and varied tit, having similar phenotypes and similar breeding habits to each other, also coexist in the ever-green broad leaved forests. The former widely distributing in palaearctic subregion has been studied well, but the latter is not sufficiently explored yet (see Higuchi, 1975, 1976), because of its narrow distribution (Formosa, Southern Korea and Japanese Islands (Ornith. Soc. Jap., 1974)). Thus, interspecific relationship between these two species in the ever-green broad leaved forests is not obvious at all. Data obtained in the present study provide new informations on the food consumptions of great tit and varied tit in the ever-green broad leaved forests. Comparisons between different habitats and between two tit spcies were attempted, and then the author discussed on two points, characteristics of the ever-green broad leaved forests as the habitat of Paridae and possible segregation between two tit species. I would like to thank Prof. Y. Ono, Department of Biology, Faculty of Science, Kyushu University, for his invaluable suggestions and laborious correction of this typescript. * Laboratory of Ecology, Department of Biology, Faculty of Science, Kyushu University, 33 Fukuoka, Japan.
2 I also very grateful to Prof. H. Kubo, Faculty of Liberal Arts, Saga University, for his continued support and encouragement, to Dr. T. Iwamoto, Faculty of Education, Miyazaki University, for helpful comments on this typescript and to the members of Laboratory of Ecology, Department of Biology, Faculty of Science, Kyushu University for field assistance. I wish to thank Mr. Y. Tsukiji, Higashisefuri Village, Saga Prefecture, who gave me accomodation to research in Mt. Gongenyama. Study area Two areas were selected for the investigation, Mt. Gongenyama, Kanzaki, Saga Prefecture, and Mt. Tachibanayama, Fukuoka, Fukuoka Prefecture. Mt. Gongenyama, 455 m above sea level, is occupied by cedar woods, bamboo forests and the ever-green broad leaved forests. Study area lies in the ever-green broad leaved forests of which altitude ranged from 250 m to 400 m, consisting of Shiia Sieboldii and Cinnamomum Camphora in the tree layer, and of Aucuba japonica, Eurya japonica, Neolitsea sericea and Quercus glauca in the shrub layer. The investigation was started in 1975 using nine nest boxes, especially deviced for automatically photographing, set up by Prof. H. Kubo, Saga University, in 1974. This kind of nest box was put on the post about 1.5 m in height and was fitted with a camera at the back wall. In 1976, additional ten nest boxes for photographing were set up by the author. Besides these nest boxes, there were about 120 nest boxes of ordinary type in the study area. The investigation was carried out from 1975 to 1977. Mt. Tachibanayama, 368 m above sea level, is situated near by the Hakata bay and is occupied by cedar woods and the ever-green broad leaved forests. The investigation was carried out in the ever-green broad leaved forests from 180 m to 250 m above sea level, consisting of Cinnamomum Camphora and Machilus Thunbergii in the tree layer and Aucuba japonica, Eurya japonica, Neolitsea sericea, Quercus glauca and Damaacanthus indicus in the shrub layer. Fourteen nest boxes for photographing were set up in 1977. Method Each food item brought to the nest was photographed with the 16 mm movie camera, 8 mm movie camera or 35 mm motor-driven camera attached at the back of the nest box. The camera was wired to the micro-switch attached at the entrance hole of the nest box. The micro-switch had a lever horizontally extending across the hole, on which a bird perched and pushed down to get into or out of the nest. A digital counter was used or direct observation was made to record the frequency of visits to the nest by parent birds for the nest box to which the camera could not be equipped. At the nest fitted with the camera, direct observation was made in order to examine the efficiency of switching. The kind and the size of food items were determined by inspecting the film with the editor-viewer or microscopic projector.
Result At both Mt. Gongenyama and Mt. Tachibanayama, great tit has only one brood and a short breeding season. The earliest date of first egg was March, 22 and the latest date was May, 24 from 1975 to 1977 (of twenty one nests). were in the nests from mid-april to mid-may. Most broods successfully fledging Broods hatching after mid-may seldom fledged successfully owing to stopping of supply of food by parents and heavy predation by the snake, particularly Elaphe conspisillata, and the weasel, Mustela sibirica itatsi. Although data on breeding are insufficient, varied tit also has only one brood and a short breeding season in both areas. The frequency of the visits to the nest by parent birds Great Tit Feeding frequency was measured at the nests of SU-3, KU-6 and NO-9 with the digital counters and automatic cameras, and U-3 by the direct observation. All nests except for NO-9, which is in Mt. Tachibanayama, are in Mt. Gongenyama. Fig. 1 and Tab. 1 show the daily change of the number of visits to the nest by parent birds. both KU-6 and SU-3 (brood size was nine and ten, respectively), the number of visits was stable at about 13 times/nestling/day except that it was relatively high (about 21 times/ nestling/day) in the middle phase of the nestling period. Differently, visits to U-3 (four nestlings) increased gradually during the nestling period. Broods of both SU-3 and KU-6 suffered from the predation by snakes and reduced to half and two-third, respectively. to each other. Responses of parents to reduction of nestlings were different SU-3 suffered from the predation and nestlings were reduced from ten to five in the afternnon of 9th nestling-day. Nevertheless, parent birds brought foods as In
Fig. 1. Daily change of feeding frequency of great tit. Open marks and x are number of visits per brood, and closed ones and cross are per nestling. Table 1. Daily change of feeding frequency (number of visits per nestling per day). In varied tit, A is number of visits and B is that of preys. mean of B/A=1.59
Food and food consumption of nestling tits 5 frequently as before for successive two days after the predation and consequently number of visits per nestling was as twice as before (Fig. 1 and Tab. 1). KU-6 suffered in the evening of 13th nestling-day and nestlings were reduced from nine to six. Parents of KU-6 adjusted their feeding efforts to reduced number of nestlings, and therefore number of visits per nestling was not changed after predation. Varied Tit Only one observation at the nest of NO-7 (five nestlings) in Mt. Tachibanayama in 1977 is available. Fig. 2 and Tab. 1 show the daily change of the number of visit to the Fig. 2. Daily change of feeding frequency of varied tit. Open circles are number of visits per brood and closed ones are that of preys brought per brood. nest and of preys brought. Feeding frequency attained to 80 times/day (16 times/nestling/ day) by 4th nestling-day and thereafter levelled off. However, varied tit often brought more than one prey at a time, 1.59 prey/visit on average (Tab. 1), so that the number of preys brought to a brood per day was about 110 preys (22 preys/nestling). The diet of nestling Great Tit The author could not collect complete data throughout the nestling period owing to predation, desertion of nest by parents and troubles in the apparatuses, e. g. the microswitch, camera and film. Consequently, data for SU-3 from 6th to 10th nestling-day, KU-6 from 9th to 15th and NO-9 for the whole nestling period (though incomplete) were obtained. Fig. 3 shows the percentage composition of the diet of the great tit nestlings. At SU-3 and KU-6 in Mt. Gongenyama, lepidopterous larvae mainly composed of Spilarctia inaequalis formed about 70% of the diet of nestlings. Besides caterpillars, lepidopterous
Fig. 3. Percentage composition of the diet of nestling great tit. Numerals above columns are number of preys. pupae, moths, spiders and diprionid larvae were taken but any of them were few. Differently from above two nests, at NO-9 in Mt. Tachibanayama, lepidopterous larvae formed only 50% of the diet of nestlings and alternatively the proportions of moths, spiders and diprionid larvae increased. Probably, Spilarctia larvae have already pupated when nestlings of NO-9 were reared, so they were not available. Fig. 4 shows the percentage distribution of weight (dry weight) of foods brought to the great tit nestlings. Mean weight of food was 79.5, 101.7 and 82.3 mg for SU-3, KU-6 and NO-9, respectively (Tab. 2). Foods weighed from 40 to 80 mg were frequently brought. Although data on the daily change of distribution of weight of foods were insufficient, a tendency is suggested that the mean weight of food gradually increased with the growth of the nestings, particularly in NO-9 nest (Fig. 4c and Tab. 2). Varied Tit Fig. 5 shows the percentage composition of the diet of varied tit nestlings. Lepidopterous larvae formed about 45% and most of them were geometrid larvae. Besides caterpillars, many moths and spiders were taken and mantids were brought occasionally. No clear trend of the daily change of the composition of the diet with the growth of nestlings were recognized. The distribution of weight of foods is shown in Fig. 6. There was no clear trend of the daily change in the weight distribution. Mean weight of food was 42.8 mg,
Food and food consumption of nestling tits Fig. 4. Weight distribution of food of nestling great tit. (a) SU-3, (b) KU-6, (c) NO-9. Histogram in black is percent distribution of caterpillars and one in white is that of other items. Fig. 5. Daily change of food composition of nestling varied tit.
Table 2. Daily change of the mean weight of prey (mg. dr.). * : Gizzard content, ( ) : Number of gizzard samples. Fig. 6. Daily change of food weight distribution of nestling varied tit. Explanation for figure is same as Fig. 4. (8)
Food and food consumption of nestling tits 9 about half of that of great tit (Tab. 2). taken. Foods weighed less than 40 mg were frequently Total amount of foods brought to the nest per day Great Tit Fig. 7 shows the daily change of the total amount of foods brought to the nest. Total amount of foods was calculated by multiplying the number of foods brought to the nest per day by the mean weight of each item. For lepidopterous larvae, mean weight was calculated each day, but mean value for other items was derived from data obtained for the whole Fig. 7. Daily change of food consumption of nestling great tit. Large circles, triangles, x and cross are total weight of foods and small circles are body weight of nestling.
observation period. Total amount of foods per nestling per day was high in the middle phase of the nestling period but it was stable at relative low level (about 1.0 g/nestling/day) in other phases. Since the value of mean weight of food just after hatching was derived from data after 9th nestling-day in KU-6, the value of total amount of foods was overstimated. According to Betts (1955) and Royama (1966), mean size (weight) of food increased with the growth of nestlings. The mean weight of food derived from gizzard contents of nestlings died in the nest just after hatching was much smaller than those in the second half of nestling period (Tab. 2). Thus, total amount of foods calculated by multiplying the number of visits with food by these values of the mean weight was much smaller (Fig. 7). Tab. 3 shows the food consumption relative to the body weight of the nestling. Water content of food was measured from specimens collected in the habitat; 80% each for caterpillars, diprionid larvae and Gryllacridae, and 65% each for moths, lepidopterous pupae Table 3. Comparison of food consumption of nestling. * : Ten nestlings hatched and decreased to five owing to the predation of snake at 10th nestling-day. ** : Nine nestlings decreased to six owing to the predation of snake at 13th nestling-day. *** : Measured at 15th nestling-day. and spiders. The values of food consumption of great tit obtained in the present study, 27-30% of body weight, were similar to those of broods of 8-9 nestlings obtained by other authors. Varied Tit Fig. 8 shows the daily change of the total amount of foods brought to the nest. The mean weight of food such as lepidopterous larvae, moths and spiders was calculated on each (10)
Food and food consumption of nestling tits 11 Fig. 8. Daily change of food consumption of nestling varied tit. Explanation for figure is same as Fig. 7. day, but those of other items were derived from data combined for 11 days of nestling period. Thus, the total amount of foods just after hatching may be overestimated. Total amount of foods per day was stable at about 0.9 g, which is 0.1-0.2 g smaller than that of great tit. Food consumption is shown also in Tab. 3 and is 19.4% of the body weight. Discussion Comparison between different Great Tit habitats Hartley (1953), Gibb (1954, 1960), Nakamura (1970) and Ogasawara (1970, 1975) investigated the preference and segregation of feeding site in Paridae in deciduous broad leaved forests, mixed woods or pure pinewoods. boles or ground in autumn and winter According to them, great tit feeds on and changes to twigs or leaves in the breeding season. This tendency in the breeding season is reflected in the diet of the nestling. In every habitat, foliage eating caterpillars formed most part of the diet, and the moths and spiders were fed rather frequently ; of. deciduous broad leaved forests (Betts, 1955, Kiuchi et al. 1970), pinewoods (Tinbergen, ever-green broad leaved forests (present study). 1960, Gibb & Betts, 1963, Won et al. 1968), and the However, there are fairly large differences in the size (weight) of food, feeding frequency and food consumption of the nestling between different habitats despite of similarity in crude figure of diet composition. Tab. 4 shows the mean weight of food in different habitats. The value obtained in the present study is remarkably greater than in any other habitats. In both Gongenyama and Tachibanayama areas, the author collected caterpillar specimens by sweeping (total twenty two 500-strokes sweeping samples) during the breeding season in 1977 and 1978. Size and weight distribution of caterpillars combining all specimens are shown in Fig. 9. Since number of sampling trials is too few and source of specimens is restricted within 2 m above (11)
Table 4. Comparison of the mean weight of prey. * : Water content is assumed as 75-80%. ** : Mean weight of a bundle of foods brought at a time. a ground, figure provides only a crude and tentative picture. The author compared this figure with that of Gibb & Betts (1963). In the ever-green broad leaved forests, caterpillars in length from 10 to 25 mm occupied most part and similar tendency is found in the English pinewoods except for slight shifting toward larger size in the figure in the present study. Moreover, nearly all caterpillars below 30 mm in length are weighed below 20 mg. dr. (80-100 mg. fr.), which, probably, is also applied to the caterpillars in the English pinewoods. English great tit prefered Panolis of which length is from 20 to 35 mm (Gibb & Betts, 1963, Fig. 5). Therefore, on an assumption that a same relationship between the size and the weight of caterpillars as in the present study is applicable to caterpillars in English pinewoods, caterpillars weighed from 20 to 80 mg. dr. were prefered most (mean = 50 mg. dr. = 200-250 mg. f r. ; closely similar to mean weight shown in Tab. 4). On the other hand, as shown in Fig. 9, great tit in the present study prefered much larger caterpillars than English tit. As refered earlier, great tit feeds in the middle layer or canopy of tree during the breeding season, sampling by (12)
Food and food consumption of nestling tits Fig. 9. Size and weight distributions of caterpillars in environment collected by sweeping, with weight distribution of caterpillars in the diet of the nestling great tit. means of sweeping near a ground might fail to reveal the availability of caterpillars to tits. Although more detailed and accurate investigation is necessary, the author refer tentatively that great tit in the ever-green broad leaved forests selected large caterpillars more strongly than in any other habitats and that this selection is resulted from abundant stocks of large caterpillars. Mean weight of food is inversely related (but not proportionaly) with the feeding frequency. In English oak wood, mean number of visits is 535/day (42/nestling/day) for large first broods (brood size>12), 450/day (50/nestling/day) for small broods (brood size< 11) and 380/day (58/nestling/day) for second broods (mean brood size= 6.8) (Gibb, 1950). In English pinewoods, early broods (hatching before June, 15) were visited more frequently than late ones (hatching after June, 16), i.e. 46/nestling/day against 30 (Gibb & Betts, 1963). According to Royama (1966), who investigated the food consumption of nestling great tit in the larch woods in Mt. Fuji, early broods (hatching in May) of 7-8 nestlings were visited for 200-500 times per day and late ones (hatching in June and July) of 6-8 nestlings were for about 150. In cases of Gibb & Betts (1963) and Royama (1966), low frequency of visits to the late broods resulted from the increase of the availability of large prey. The number of visits in the present study is smaller and, as refered earlier, mean weight of food is much heavier than those in any other areas. Probably, low frequency of
visits in the present study also arised from the high availability of large prey. As shown in Tab. 3, amounts of food brought per nestling per day and the daily food consumption are similar between different habitats, though differences are fairly large in the mean weight of food and feeding frequency (Tabs 1, 2 and 4). However, in the cases of Gibb & Betts (1963) and Royama (1966), amount of food and food consumption differed between broods of different size. Values of the broods of 5 6 nestlings (reduced owing to the predation of snakes) in the present study are closely similar those of the broods of 8-9 nestlings in other areas. Since they were not undernourished (2 g heavier in the body weight than those of Royama (1966) of the same sub-species), nestlings in the preset study were reared healthy with smaller amount of foods fed than in any other areas. Royama (1966) pointed out the heat loss from the body as the factor influencing the food consumption of the nestling and refered that the food consumption of the nestlings in the cold and wet environment was greater. Small food consumption in the present study could be explained that the nestlings were in the warmer environment than in any other areas. Varied Tit Nakamura (1970) and Higuchi (1975) refered that varied tit frequently ever-green broad leaved forests and foraged in the upper and middle layer of trees. occurred in the Moreover, varied tit often forages insects resting on the leaves or flying in the air by hovering or fly-catching (Higuchi, 1975). Presumably, because of those foraging habits, moths and spiders occupied greater proportion in the diet of nestlings compared with great tit, though caterpillars still formed about half of diet. areas (cf. Won et al. 1965, Higuchi, 1975). The diet composition is similar between different However, Won et al. (1965) observed that the proportion of spiders in the diet consistently decreased with the growth of nestlings. Similar tendency was observed in great tit (Royama, 1970) and coal tit (Haneda & Horiuchi, 1970). Royama (1970) stressed the role of spiders as the source of nutrient necessary for nestlings in particular stage. However, the proportion of spiders did not change in the present study. Concerning to feeding frequency and food consumption of nestling varied tit, there are few comparable investigations. According to Won et al. (1965), a brood (not described about brood size) was visited 28 times per day just after hatching and 139 times on 20th nestling-day (just before fledging). The latter is about two-fold of one obtained in the present study. However, mean weight of food is uncertain in the study of Won et al. (1965), so that further comparison is impossible. Comparison between great tit and varied tit As refered earlier, great tit forages in higher layer of trees in the breeding season than in other season, which slightly facilitates overlapping of foraging site with varied tit. However, the degrees of segregation is still large. For example, there were few prey species commonly eaten by broods of both NO-7 (varied tit) and NO-9 (great tit) which (14)
Food and food consumption of nestling tits 15 were reared in the same period in a same area. in prey species but in the size of the prey; Difference of preference is found not only great tit fed much larger prey than varied tit (Tab. 2). Betts (1955) and Gibb & Betts (1963) found the correlation between the size of the prey prefered and the shape of breaks and the size of the body in English four tit species coexisting in a same habitat. In the present study, no attempt was done in measuring of every part of the body of adult tits, but, in general, varied tit has somewhat larger beaks and body than great tit (Kobayashi, 1967). Nevertheless, the former prefered smaller preys. Hunting method may be related to the size of the prey prefered. Varied tit brings more than one prey at a time, which may be related with prefering smaller prey. Mean weight of foods per visit is similar to those of great tit in the other area, but is still smaller than that of great tit in the present study (Tab. 4). The food consumption of varied tit nestling was much smaller than that of great tit, because of high proportion of moths and spiders in the diet, the low feeding frequency and the small mean weight of prey. Since the body weight of nestling just before fledging was similar to that of the same sub-species (P. v. varius) in other area (Higuchi, 1975), varied tit nestlings in the present study were not undernourished. Royama (1966) stated that the daily food consumption relative to the body weight was similar between different species with different body weight. However, the difference of the daily food consumption between great tit and varied tit is fairly large in the present study. In this paper, the author cannot make conclusion which result is correct, because the comparison is only based on dry weight of foods. The conclusion will be given implicitly when more field data about the food consumption of nestling varied tit are collected and foods of both tits are analyzed calorimetrically and nutritionally. Summary : 1) Feeding rate and food consumption of nestling tits, great tit and varied tit, were assessed by means of automatically photographing of foods brought to the nest, in the ever-green broad leaved forests in northern Kyushu. 2) In the diet of great tit nestling, lepidopterous larvae formed about 70%, and moths, spiders and diprionid larvae are other important preys. Mean weight of prey is 87.0 mg, dr. and preys weighed from 40 to 80 mg. dr. were brought most frequently. In the varied tit nestling, lepidopterous larvae formed 45% and, alternatively, proportion of moths and spiders increased. Mean weight is 42.8 mg. dr. and preys below 40 mg. dr. were brought frequently. 3) Feeding frequency of great tit was stable at about 13/nestling/day except that it was at about 21/nestling/day in the middle phase of nestling period. Varied tit often brought more than one prey at a time, 1.59 preys/visit on average. Feeding frequency of varied tit was stable at 14/nestling/day (22 preys/nestling/day) after 4th nestling-day, though it was somewhat lower just after hatching. 4) Daily amount of foods just before fledging was 4.67g. fr./nestling/day and 29.0% of the nestling body weight of great tit. For varied tit, 3.16 g. fr./nestling/day and 19.4%. 5) Diet composition, feeding rate and -food consumption are compared between several different habitats and between above two species, and was discussed. (15)
Appendix Royama (1966) estimated the amount of foods brought to the nest of great tit from the size of the prey measured by the direct observation. On an assumption that a prey has cylindrical body shape, volume index is calculated by multiplying the square of the body diameter by the body length. Then, the body weight (dry weight) is plotted against the volume index and a linear relationship is obtained. He applied this method to the estimation of body weight of caterpillars and moths. In the present study, two methods were attempted. One is the body diameter-body weight regression method (D-W method) and another is the volume index-body weight regression method (V-W method). Regression lines were obtained separately with following items; geometrid larvae, other lepidopterous larvae, lepidopterous pupae, moths, spiders, diprionid larvae and Orthoptera (mainly Gryllacridae). However, the body weight of the mantid was estimated by the head width-body weight regression and the abdomen width-body weight regression. There are fairly large differences between values of body weight estimated by different two methods, particularly in geometrid larvae the difference is remarkable (Tab. A). Since the bird Table A. Comparison between values of weight of each food item estimated by D-W method and V-W method. often brings the prey holding in its bill on one end of the prey, the body of the prey which has soft abdomen is stretched. Hence, the body length may be measured greater than real value, which may result in overestimating in the body weight of the prey. Therefore, the value estimated by D-W method has smaller bias than by V-W method. Besides this, in varied tit there were few samples completely measured of the body length. Because of above two reasons, the author have estimated the body weight only by D-W method. References Betts, M. M. 1955. The food of titmice in oak woodland. J. Anim. Ecol. 24: 282-323. Gibb, J. 1950. The breeding biology of the great tit and blue titmice. Ibis 92: 507-539. Gibb, J. 1954. Feeding ecology of tits, with notes on tree-creeper and goldcrest. Ibis 96: 513-543. Gibb, J. 1960. Populations of tits and goldcrests and their food supply in pine plantations. Ibis 102: 163-208. Gibb, J. A. & M. M. Betts, 1963. Food and food supply of nestling tits (Paridae) in Breckland (16)
Food and food consumption of nestling tits 17 pine. J. Anim. Ecol. 32: 489-533. Haneda, K. & T. Horiuchi, 1970*. Food and food-consumption of nestling coal tits Parus ater. Bull. Inst. Nat. Educ. Shiga heights 9: 31-43. Hartley, P. H. T. 1953. An ecological study of the feeding habits of English titmice. J. Anim. Ecol. 22: 261-288. Higuchi, H. 1975*. Comparative feeding ecology of two geographical forms of the varied tit, Parus varius varius in Southern Izu Peninsula and P. v. owstoni in Miyake I. of the Izu Is. Tori 24: 15-28. Higuchi, H. 1976. Comparative study on the breeding of Mainland and Island subspecies of the varied tit, Parus varius. Tori 25: 11-20. Kiuchi, K., C. Matsuzawa, K. Noda, H. Satoh, H. Hosoda, S. Maruyama & T. Tawara, 1970*. Nestling food of some birds in coniferous forest. I. List of food items. Bull. Inst. Nat. Educ. Shiga heights 9: 45-58. Kobayashi, K. 1967**. Birds of Japan in natural colours. 229 pp. Hoikusha, Tokyo. Nakamura, T. 1970*. A study of Paridae community in Japan. II. Ecological separation of feeding sites and foods. Misc. Rep. Yamashina Inst. Ornith. 6: 141-169. Ogasawara, K. 1970*. Analysis of mixed flocks of tits in the Botanical Garden of Tohoku University, Sendai. II. Foraging layers by species and their interrelations within the mixed flock. Misc. Rep. Yamashina Inst. Ornith. 6: 170-178. Ogasawara, K. 1974*. Analysis of mixed flocks of tits in the Botanical Garden of Tohoku University, Sendai. IV. Foraging habits and supplanting attacks among species forming mixed flocks. Misc. Rep. Yamashina Inst. Ornith. 7: 637-651 Ornith. Soc. Jap. 1974. Checklist of Japanese birds (2nd printing). Gakken, Tokyo. Royama, T. 1966. Factors governing feeding rate, food requirement and brood size of nestling great tits Parus major. Ibis 108: 313-347. Royama, T. 1970. Factors governing the hunting behaviour and selection of food by the great tit (Parus major L.). J. Anim. Ecol. 39: 619-666. Tinbergen, L. 1960. The natural control of insects in pinewoods. I. Factors influencing the intensity of predation by songbirds. Arch. Neerl. Zool. 13: 265-343. Won, P., S. Kim & C. Kim, 1965*. Breeding biology and chick food of varied tit Parus varius. Misc. Rep. Yamashina Inst. Ornith. 4: 198-207. Won, P., H. Woo, K. Ham & M. Chun, 1968*. Chick food analysis of some Korean birds (III). Misc. Rep. Yamashina Inst. Ornith. 5: 363-369. * in Japanese with English summary. ** in Japanese.