The Embryology of Chlamydoselachus 627 



plate I), and are then or later thrown out into the sea. Two investigators (Hawkes, 1907; 

 and Smith, 1937), have found highly vascularized areas in the wall of the right uterus. 

 These observations suggest that these areas might have served to secrete food stuffs into 

 the uterus. Then the long gestation period and the enormous size of the relatively late 

 embryos still attached to large yolk sacs seem to indicate that these embryos grow not 

 at the expense of the yolk alone. All these things lead to the inevitable question — ''Do 

 the external gills of the embryos of Chlamydoselachus serve to absorb food from a uterine 

 secretion?" The facts and inferences as to such a possible source of food in Chlamydo^ 

 selachus have been set forth above. It seems quite sure that in any case, these external 

 gills of the non^extruded juvenile sharks serve as respiratory organs. 



Chlamydoselachus has been ranked by the systematists as the lowest, most primitive 

 living shark. Yet in its reproductive organs, as this paper shows, and in many other 

 organs, as Smith has pointed out in his monograph on the anatomy (1937), it is very 

 highly specialized. Its embryos have external gill-filaments, which never grow very long 

 and which eventually shorten until they are almost or quite concealed from view by the 

 gill'flaps. Now external gills, as they are ordinarily understood, are embryonic or an- 

 cestral organs which tend to become eliminated in the process of evolution. External 

 gill-filaments are either evanescent structures of external origin, developed as outgrowths 

 on the outermost edge of a visceral arch before the clefts have broken through, or they 

 are precocious growths of normal gills which tend to shorten in the course of later de- 

 velopment. Let us now study these filaments as they are shown in the drawings of the 

 embryos of this archaic shark and see to which category they belong. 



The first evidence of the presence of gill-filaments in the embryos of Chlaynydo- 

 selachus is found in the 25-mm. stage as portrayed by Ziegler (1908, Fig. 2). Ziegler's 

 figure (my Text-figure 27c) is poorly reproduced on soft paper, and the budding filaments 

 show up indistinctly. However, in the figure these buds appear not on the outer edges of 

 the gill-arches but on the hinder inner sides of arches 1-5, and on both sides of No. 6. 

 Furthermore, it is clear that the gill-clefts have become perforated, and that the outside 

 liquid penetrates into the pharynx through the slits. 



Brohmer's drawing (1909, Fig. 3) of his 25-mm. specimen (also poorly reproduced) 

 shows the gill-filament buds on both sides of arch No. 2 (my Text-figure 27a). Probably 

 they are present in a beginning stage on both sides of each arch. They are next seen (as 

 far as data are at hand) in Nishikawa's 32-mm. embryo (Figures 19-21, plate II) where 

 they appear on both sides of every arch — excepting of course the first. Passing over the 

 34-mm. specimen, we go to the 39-mm. embryo in which stage the filaments first begin to 

 show externally, particularly in the first sHt (Figure 25, plate II). These filaments are in 

 about the same stage of development as those in Scammon's 18-mm. Squalus (1911, Fig. 

 27, pi. III). But in size and abundance they are far behind the filaments protruding from 

 the gill-sHts of Scammon's 20.6-mm. embryo of the dogfish. 



External filaments are seen to be pretty well developed in Dean's ''new series" of 

 drawings of embryos 46, 54, 66, and 103 mm. in length. These can be best studied in the 



