654 BIRDS 



In the snakes, we have a pretty analogy for the situation in the tele- 

 osts, for here a mesodermal vascular papilla has been given a trial, and 

 practically abandoned in favor of a vitreal-vessel system (the latter pre- 

 sumably more efficient, since it is in so immediate contact with the tissue 

 which it serves). The retinal plexus does not seem to have reached a 

 high state of development in the Boidse (which are nocturnal), but it 

 has full expression in the Colubridae, where the total area of its vessels 

 is said to equal one-third of the whole area of the retina; and it has 

 persisted unchanged in the higher families despite their wholesale re- 

 versions to nocturnality — perhaps because the ophidian chorioid had 

 become so very thin, so that it was as easy to keep the hyaloid vessels as 

 to discard them and rebuild the chorioid (cf. Protopterus!) . 



To anticipate the next Chapter : the mammals characteristically have 

 many vessels and capillaries embedded in the inner layers of the retina. 

 This greater intimacy of relationship, as compared with the fish-anuran- 

 snake situation, is only to be expected since the mammals are warm- 

 blooded and those other groups are not. Just so, the buttressing or 

 pleating of the avian pecten (often claimed to promote structural 

 rigidity, which of course it incidentally does) is a secretory-surface- 

 increasing device which these hot-blooded creatures require, in contrast 

 to the lizards and the extinct reptiles which really evolved the lizard 

 conus. A vestigial conus occurs in many of the lower mammals, partic- 

 ularly in marsupials and rodents. The retinal vessels are lacking in the 

 monotremes, and are lacking or greatly reduced in many other nocturnal 

 mammals. They are best developed in the (diurnal) primates and in the 

 (arhythmic) ungulates and carnivores. Where, as occasionally, the 

 retinal vessels extend out to the outer nuclear layer as if to supply even 

 the visual cells themselves, it is in forms whose chorioids are exception- 

 ally under-developed (dormice, flying-squirrels) or are insulated from 

 the visual cells by a relatively impermeable retinal tapetum (opossum; 

 the retinal tapetum of the crocodilians also seems to interfere, for the 

 chorioid is extra-thick behind it — p. 613). It is a toss-up whether the 

 embedded retinal-vessel system of mammals, or the pecten of the birds, 

 is 'better'. The retinal vessels are a more direct means of supplying the 

 retina; but the pecten perhaps interferes less with vision — the mam- 

 malian retinal vessels have always to be excluded from the vicinity of a 

 fovea. 



The whole s N d picture thus reveals a rather consistent relationship 

 with habits which would seem to carry with them a high level of retinal 



