EARLY HISTORY OF THE PLACENTALIAN EYE 687 



of a former universal nocturnality — perhaps even more complete than 

 that of any monotremes, Hving or dead, since even the scleral cartilage 

 has not been kept. In the opossums, which are the most archaic of living 

 marsupials and hence, so to say, have had the most time in which to 

 get rid of useless structures, some of the single cones have lost their 

 oil-droplets. 



The placental mammals must have gone farther in adaptation for 

 dim-light activity, early in their history, than the marsupials have ever 

 done. Their eyes are in fact best understood not by comparison with 

 those of the lower mammals, but by comparison with those of the snakes. 

 The early snakes so completely lost the reptilian assortment of special 

 ocular structures that when the snake eye was rebuilt, upon the return 

 of the snakes to the earth's surface, it ended up as a spherical organ with 

 an entirely fibrous wall, with the lens and ciliary body out of contact 

 (necessitating a new and special method of accommodation), with a 

 wholly new set of visual cells, and (eventually) a yellow lens as a sub- 

 stitute for the ancestral diumal-lacertilian yellow oil-droplets. To a 

 degree, the placentalian eye incorporates equivalent changes and sub- 

 stitutions. The 'original' placentalian eye was of course not really degen- 

 erate like that of a mole or mole-rat, but it did take several steps down 

 the same path which the eye of the incipient snake followed to its 

 bitter end. 



Whether the placentals evolved directly from nocturnal marsupials, 

 or turned nocturnal after a derivation from diurnal common ancestors 

 of the modern marsupials and the placentals, we cannot know; nor 

 would the knowledge have much importance. We can be sure that at 

 an early period in placentalian evolution, the only placentals on earth 

 were so thoroughly nocturnal that their eyes had no stiffening structures 

 to keep them from being spherical, had large pupils and large, simple 

 lenses with no trace of a ringwulst and no contact with the ciliary body, 

 had rudimented intra-ocular muscles which were unstriated and did not 

 include a dilatator pupillae, and had no accommodation whatever. 



Now, what was the retina like in these strictly nocturnal, 'bottle-neck' 

 insectivores? Apparently all of the lowest living orders of placental 

 mammals have pure-rod retinae. But the higher ones have both rods and 

 cones. Do the cones of the higher placentals represent sauropsidan- 

 monotreme-marsupial cones which squeezed through the primitive in- 

 sectivoran knot-hole, or are they somehow new? 



