I 



TROPICAL CLIMATES AND BIOLOGY — CARTER 439 



species cooling of the water by the rain. In the laboratory the easiest 

 ■way to induce many of the frogs and fishes to lay eggs is to sprinkle 

 the aquarium with cool water. 



It is far less easy to see how biological rhythms are controlled in 

 parts of the Tropics where seasonal changes of the climate are slight. 

 We have seen that in equatorial regions the only large seasonal change 

 is in rainfall, but it does not seem likely that variations in rainfall 

 are the effective control of the rhytlims, for there is rain in every 

 month of the year, the humidity is always high and not significantly 

 variable, and, though food may for some animals differ from month 

 to month, it is always plentiful. Yet the fact is that most species 

 have well-defined seasonal rhythms even in these environments, 

 though a few breed all the year round and some others have double 

 breeding seasons associated with the double seasonal change. Baker 

 and his coworkers found [9], for instance, that in the rain forests at 

 Noumea in the New Hebrides, a highly invariable climate though the 

 latitude is 15° S., all the species he worked on were seasonal in their 

 breeding, the birds and mammals at least as markedly seasonal as in 

 temperate comitries. A lizard {Emoia sp.) had a less clearly de- 

 fined breeding season, though even it showed a seasonal rhythm of 

 gonad growth. A bat [Miniopterus sp.) , which spent the day in caves 

 where the climate was even more invariable than in the forest outside 

 the caves, was the most markedly seasonal of all, breeding on only a 

 few days at the beginning of September. The breeding seasons were 

 often not the same as those general in temperate regions, and in 

 the case of a passerine bird, Pachycephala pectoralis, differed from its 

 breeding times in places at the same latitude in Australia. He had 

 evidence that the times of breeding persisted at the same dates from 

 year to year. He was not able to find any seasonal climatic change that 

 could control the periodicity of the animals. Owing to the latitude 

 the length of day at Noumea varies by 1% hours, but he concluded 

 that this was not the effective cause. 



A large majority of species are seasonal in other invariable en^^ron- 

 ments both in breeding and migration. Marshall and Williams found 

 [10] that at Entebbe in Uganda on the Equator the yellow wagtail 

 {Motacilla fava), which "winters" in Africa and spends the summer 

 in Europe, was seasonal in the development of its gonad during its 

 tune in Uganda (December to April) . They were imable to find any 

 climatic change during this time to account for the periodicity. It 

 is certainly not controlled by the length of day, for this does not vary 

 significantly at Entebbe between December and April. 



For some species details of their habits and biology provide the 

 answer. Birds that nest on islands in rivers, or on river banks near 

 the water, may be able to breed only in the drier montlis when their 



