584 HIGHER FISHES 



tomical name which expresses its function (?) is 'tensor chorioideae'. 

 Though it is absent in those teleosts which have no accommodation, it 

 is not directly concerned in that process — rather, it seems to serve to 

 tauten the chorioid and retina around the vitreous body, thus preventing 

 the backward-moving lens from using the vitreous to push the retina 

 backward, which would defeat the accommodatory purpose of the re- 

 traction of the lens. This function of the muscle has never been estab- 

 lished experimentally, however; and, strictly, its usefulness must be 

 regarded as unknown. 



In the ciliary zone, the chorioid merges imperceptibly into the true 

 iris, which is fairly complex in structure. Typically, there is an anterior 

 layer of mesothelium, continuous by way of the annular ligament with 

 that of the cornea. Behind this is a thick argentea continuing that of the 

 chorioid. The pigmented and very richly vascular stroma posterior to 

 the argentea bears, superficially, a scattered layer of chromatophore cell- 

 bodies whose processes perforate the argentea and expand within it, or 

 more often anterior to it (beneath the mesothelium) to contribute to the 

 externally-visible color pattern of the iris as a whole. The anteriormost 

 of the epithelial retinal layers of the iris is always heavily pigmented 

 except toward the pupil where it is converted into the lightly pigmented 

 (sometimes unpigmented) 'sphincter'. The posterior retinal layer is pig- 

 mented only in the pupillary half of the iris, and is blank behind the 

 annular ligament as in the ciliary zone. Between the stroma and the 

 retinal layers there is a conspicuous membrane which gives one the im- 

 pression of a myoid dilatator-sheet (as in the mammals — see Fig. 7b 

 and g, p. 15) ; but this membrane is only a basement-membrane, com- 

 parable with the glass membrane in the region of the sensory retina — 

 any dilatator elements ever present in teleosts are pigmented spindle- 

 shaped cells, detached from the epithelium and lying in the stroma. The 

 sphincter, when present, is not so well separated from the generative 

 epithelium as it is in mammals. Very often a 'sphincter', sometimes a 

 massive one, is present without demonstrable contractility. The vascular- 

 ization of the iris is complex and variable, and its different plans in 

 different groups have yet to be fully interpreted and unified; but the 

 uncertainties here, in this blind-alley group of vertebrates, are of no 

 consequence to the phylogenetic theme of these synoptic chapters. 



The Teleost Retina — In so huge and diversified a class of vertebrates, 

 the retina naturally shows great differences from one group to another. 

 The fishes of the caves and crevices, muddy waters, and the deep sea 



