(140) OBSERVATIONS ON SWALLOWS AND HOUSE- MARTINS AT THE NEST. BY R. E. MOREAU AND W. M. MOREAU. RECENT studies of the parental care by African Hinindinidae and Swifts have suggested that, in addition to specific differences in the share taken by the sexes, certain rhythms of the behaviour may be specific, especially the favourite duration of the individual spell on the eggs, the duration of the intervals when the eggs are left uncovered and the frequency with which food is brought to the young (Moreau, 1939 (a), and in Press). During our leave in this country we hoped to make for comparison some preliminary observations on the British Hirundinidse. Owing to bad weather and other circumstances beyond our control we have been able to do much less than we had hoped; but so little has been published on the numerical aspect of these birds' behaviour that it seems worth while to put some of our information on record. Some notes on the Sand-Martin have already appeared (Moreau, 1939 (b)). Taken in conjunction with the African data, those on the British species certainly suggest the possibility of specific differences that would repay further investigation. Our observations were made near Farnham, Surrey. Especially in connection with those on the House-Martins we have to thank Mr. W. F. Rankine for facilities and co-operation. INCUBATION. SWALLOW (Hirundo r. rustica). The information in The Handbook of British Birds is that incubation is " apparently by female only or share of male is small." On eleventh and twelfth days of incubation (July 2ist and 22nd) we watched a Swallow's nest from 4 p.m. to 6.50 p.m. and 2.15 p.m. to 5.15 p.m. respectively. On the first day the eggs were brooded 9 times by the female, 4 times by the male ; on the second day 6 times by each bird. Individual spells on the eggs lasted from 5 to 16 minutes (with no obvious sexual difference) ; and of the 21 complete spells recorded, 11 were of durations of 8 to 12 minutes. No instantaneous change-overs took place. The eggs were left uncovered for intervals of 2 to 9 minutes, more than half of the intervals lasting 5 to 6 minutes. There was no regular alternation between male and female ; and on one occasion both returned at once, the male settling on the eggs. (As a
VOL. xxxiii] OBSERVATIONS ON SWALLOWS. 147 rule the male alighted for a short time up to 2 minutes on a neighbouring beam before going on the nest, while the female regularly flew straight to the nest from outside the building.) Both afternoons were cool, the shade temperatures not exceeding 18 0 C. (65 0 F.); the eggs were, however, covered for only 64 per cent, and 65 per cent, of the observed time. HOUSE-MARTIN (Delichon u. urbica). Thirty yards away from the Swallow's nest just referred to, two House-Martins' nests, a foot apart, were under observation : Nest A : June 24th, 2 p.m. to 6 p.m.; 25th, 10 a.m. to 1 p.m. (About the eighth and ninth days of incubation.) Nest B : July 10th, 3.40 to 7 p.m. (About the seventh day of incubation.) Apparently an important characteristic of the House- Martin's incubation is the instantaneous change-over by the parents. As a rule what the observer saw was a bird entering the nest and after a short interval (up to 45 seconds) a bird emerging. It was (and would be at most nests) usually impossible to be certain that the emergent bird was not the one that had entered just before; but on a few occasions we were able to exclude this possibility and to see with certainty that the sitting bird came off and was not merely visited by its mate. Provisionally, then, we conclude that each entrance followed by an exit at nests A and B represented a change-over on the eggs. At nest A 21 of the 28 spells "on", that we recorded, were terminated by this sort of change-over, and at nest B all the recorded spells. None of the intervals when the eggs in nest A were uncovered exceeded 4 minutes ; so that in the two nests the eggs were covered 95 to 100 per cent, of the observed time. At the two nests together 41 complete spells " on " were recorded, varying in duration from 4 to 24 minutes, with a strong preference for durations 7 to 15. FEEDING OF YOUNG. SWALLOW. Four nests, A, B, C, D, each with 3 young (an abnormally small number), on rafters immediately under the slates of a timber shed, were under observation at various dates in July. Another nest, E, with 4 young, on a beam in a cowshed, was watched on July 10th. Owing to the striking effect of rain on feeding (see below) the following data on feeding-rates relate to rain-free periods
148 BRITISH BIRDS. [VOL. xxxm. only. The ages given for the young are estimated on the assumption that on the dates they flew (which are known) they were 21 days old, the usual fledging age according to The Handbook of British Birds. - Nests B and C were observed concurrently. TABLE 1. Nest. No. of Estimated No. of feeds brought in separate complete Young. Age in days. ' rain-free hours. A 3 3 24, 20, 24, 27 4 23 20 21, 25 B 3 11 16, 19 C 3 15 13, 14 E 4 18 38, 39, 38 (see below) The consistent nature of the feeding during rain-free periods is shown by the fact that of the 340 intervals between feeds recorded, only 32 exceeded 5 minutes and only 5 exceeded 9 minutes. These lasted 11, 11, 12, 13 and 14 minutes. Owing to the dimly lit situations of the nests and the rapidity of the visits with food it was difficult to assess the shares of male and female. At nest E, where it was least difficult, it appeared that rather less than half the visits were made by the male and that a second female was visiting the nest, at least occasionally. More than once a bird (? sex) approaching the nest before the food-bringer had departed was chased away by the latter. Subsequently the male, making a visit to the nest, was followed by a female, which he drove away before feeding the young. While he was still busy two birds (? sex) flew in and out chasing each other past the nest. Six minutes later two apparently adult females arrived at the nest together and both fed the young.* The feeding-rate at this nest was relatively high, even allowing for the fact that it contained one more young bird than the others; and this might to some extent be accounted for by the participation of the third " parent." (The date being July 10th it is impossible that this could have been a member of an earlier 1939 brood.) The effects of rain are shown by the following details:. On July 4th heavy rain fell 10.13 to 11.10 a.m. At nest A, with young 4 days old, the parents remained active and Visited the nest 19 times during the 57 minutes nearly as often as in the rain-free hours (Table 1). At 10.29 a.m. one Emmet (1939) has observed that " Swallows other than the parents paid visits to the nest." He has recorded 48 feeds brought to 4 young in a single hour.
VOL. xxxin.] OBSERVATIONS ON SWALLOWS, 149 bird brought a fair-sized moth, but did not try to give it to the young ; and after sitting near the nest with it for a few minutes it carried it out of the building again. On July 6th light rain began at 2.1 p.m. and from 2.18 onwards continued more or less heavily until the end of the watch at 4 p.m. At nest A, with young 6 days old, 9 feeds were brought up till 2.46 p.m., and only 2 more (at 3 p.m. and 3.46 p.m.) during the remaining 74 minutes. (But the young birds were brooded more than during the rain-free periods.)* At nest B (observed concurrently with A), with young 11 days old, 14 feeds were brought 2.1 p.m. to 2.36 p.m., none 2.36 p.m. to 3.36 p.m. and four 3.36 p.m. to 4 p.m. At nest C (observed concurrently with B and A), with young 15 days old, 13 feeds were brought 2.1 p.m. to 2.37 p.m., none 2.37 p.m. to 3.18 p.m. and five 3.18 p.m. to 4 p.m. On July 20th very heavy rain fell 10.55 a- 111 - to 11.18 a.m. At nest A, where the young were now about 20 days old, feeding, which both before and after was at over twenty times an hour, stopped entirely 10.56 a.m. to 11.27 a - m - From the foregoing examples it appears that heavy rain practically stops the feeding of the young. On July 6th there was a lag of about 15 minutes after heavy rain began before feeding was seriously affected, but then feeding stopped at nests B and C almost simultaneously. Except at nest A on July 20th we had no evidence that parents with grown young came in to shelter near the nest. The amount of attention given to the young at nest D was in striking contrast to that at the other four Swallows' nests. Nest D was watched from 1.25 p.m. to 5.15 p.m. on July 3rd, and, concurrently with nest A, from 9.32 a.m. to 12.26 p.m. and July 4th. On the 5th, or early on the 6th, the young left the nest. The parents at the other four nests fed their young very regularly and practically never approached the nest without giving food. The parents at nest D entered the building fairly often, either singly or together, but as a rule either settled near the nest or fluttered for a few seconds a short distance from the clamouring young. During these " feints " the old birds sang a good deal; and several times after a " feint " or We formed the opinion that only the female actually brooded, though the male often sat for a few minutes on the edge of the nest. This is what happens with Hirundo smithii. But nest A could not be seen clearly enough for our observations to be as precise as we should wish.
150 BRITISH BIRDS. [VOL.XXXIII. an actual feed the parents sat for a short time on a beam near the nest and then " feinted " again. On July 3id there were 13 visits 1.25 p.m. to 3.16 p.m., 8 of them certainly feints, followed by 10 feeds 3.17 p.m. to 3.34 p.m., 6 feints 3.35 p.m. to 4.24 p.m., 21 feeds and 4 feints 4.25 p.m. to 5,2 p.m. On the 4th, during the period 9.32 a.m. to 11.25 a.m., which included the rainy spell 10.13 a.m. to n.10 a.m. (not inhibiting activity at nest A), the parents at D made only 5 visits, all feints, and no visits at all 12.3 p.m. to 12.26 p.m., but brought 25 feeds in the 37 minutes 11.26 a.m. to 12.3 p.m. The young birds between them dropped at least four fecal pellets during the " hungry spell" 9.32 a.m. and 11.25 a.m., so they had evidently had a number of feeds early in the morning before observation started. Since these young birds were practically ready to fly we take it that the parents' behaviour may have been intended to hurry them out of the nest, partly by positive persuasion and partly by making the food-supply irregular. We have some evidence of similar procedure, but not so marked, at African nests of some Hirundinidae. It is to be noted that the behaviour of the parents at nest A when the young appeared to be practically fledged, did not resemble that at D. HOUSE-MARTIN. To a nest with three young about ten or eleven days old 46, 39, 42 and 36 feeds were brought in 4 separate complete rain-free hours observed. The feeding was very regular ; only three intervals exceeded 4 minutes and those lasted 5, 7 and 8 minutes. The young first left the nest on the morning of July 21st, when they were all out on surrounding roofs. In the afternoon there was much movement in and out of the nest, but most of the time two were in and one was out on a roof, all receiving frequent feeds. The following morning they mostly spent out again, according to Mr. Rankine's information, but all three passed practically the whole afternoon and evening in the nest, where in complete hours they received 31 and 32 feeds. Incidentally this is a good example of how easily a deceptive fledging period might be obtained for a nest. SUMMARY. Data are derived from about 45 and 20 hours' observations on Swallows' and on House-Martins' nests respectively. At a Swallow's nest incubation was shared fairly equally by the parents, with a preference for individual durations " on " of 8 to 12 minutes and " off " of 5 to 6. House-Martins' eggs at
VOL. xxxni.] OBSERVATIONS ON SWALLOWS. 151 two nests were covered much more consistently, the sitter very rarely coming off until relieved. The usual spell " on " lasted 7 to 14 minutes. In the Swallow three broods of three received on the average 21 feeds an hour when no rain fell, compared with over 30 at a House-Martin's and at Sand- Martins' nests, all with three young. A brood of four Swallows averaged 38 feeds an hour, but apparently a third adult (a female) was helping. One brood of Swallows nearly ready to fly was fed very irregularly and most visits were " feints," apparently to coax them out. Except at this nest food was brought most regularly during rain-free periods, hardly any intervals exceeding 5 minutes. Heavy rain usually stopped feeding after a lag (15 to 20 minutes). REFERENCES. EMMET, H. J. 1939. " A diary of Swallows." Bird Notes and News, 18, 145-8. MOREAU, R. E. (1939) (a). "Parental care by some African Swallows and Swifts." Bull. Brit. Orn, Club, 59, 145-9. MOREAU, R. E., in Press. " Numerical data on African Birds' behaviour at the nest." I. Hirundo smithii. II. Psalidoprocne holome- Imna. Proc. Zool. Soe. London and Ibis. MOREAU, R. E. and MOREAU, W. M. (1939) (b). " Observations on Sand-Martins at the nest." Brit. Birds, XXXII, 95-7.