LARVAL OR SWIMMING STAGES 25 



10, 12, 13, which correspond with the figures by Hatschek for Teredo, by 

 Goette and by Schierholz for Anodonta, and by Ziegler for Cyclas. 



An organ upon the origin of which renewed observation should be 

 made in the American oyster is the shell-gland. Brooks described the 

 shell-valves as being formed apart, at the ends of a crescent-shaped trans- 

 verse groove (liis fig. 36; my Plate V, fig. 24), which is prolonged (deepened) 

 at its centre into the primitive digestive cavity at the blastopore. Shortly 

 afterwards Horst discovered in the European oyster that there is a second 

 invaginated depression of the ectoderm (his fig. 8; my fig. 20), distinct 

 from, smaller and more temporary than that of the blastopore, situated on 

 the opposite side, near the original animal pole. He expressed the opinion 

 that what Brooks had taken for the blastopore was nothing but the open- 

 ing of this preconchylian gland. 



That Brooks made some mistake is certain, not only from his association of the 

 shell- valves with the blastopore, but from his location of the permanent mouth, and his 

 orientation of the larva with respect to dorsi-ventrality. His careful continuous ob- 

 servations on the younger stages preclude the probability of his misunderstanding the 

 blastopore and its origin. It is evident from the lack of reference to it, that what he 

 did not comprehend was the shell-gland, which originates about the time of closure of 

 the blastopore. Taking into account the difficulty of turning without damaging the 

 young larva in a microscopic preparation, so as to see the surface of the same specimen 

 from all directions, the depression of the shell-gland might easily be overlooked or mis- 

 taken for the blastopore. 



According to Horst, the shell-gland originates (Plate V, figs. 20, 21) as a distinct 

 depression, a slight invagination of the ectoderm cells, on the upper surface, just behind 

 the dorsal pole, and opening across the long axis. It deepens, its cells become high and 

 cylindrical, its mouth narrows. Later it loses its original character as an invagination, 

 and forms an ectodermic thickening that secretes a thin cuticular membrane, the first 

 indication of the shell. Carbonate of lime is not deposited in it until later. He agreed 

 with Davaine that the hinge-region is first produced and that the shell originates as a 

 single piece, contrary to Lacaze-Duthiers and Brooks, who believed that the two valves 

 are at first separate. Horst also remarks that "the ectodermic cells, which are found 

 on the surface of the shell, have become very thin and transparent, so that the outlines 

 can not be distinguisded, but only the refringent nuclei." This is a remark difiicult to 

 understand and to correlate with his statement about "an ectodermic thickening that 

 secretes a thin cuticular membrane," for how could cells be found on the surface of a 

 cuticular membrane or shell? It must be remembered that older embryologists were 

 frequently content with few, isolated, disconnected observations, and that, therefore, 

 summary statements that are placed together may refer to stages that, in an extensive 

 series of continuous observations, would be separated. It seems possible that the case 

 where the shell-gland "loses its original character of an invagination and forms an ecto- 

 dermic thickening" was one of evagination of the shell-gland, if not an abnormality. 

 In any case it raises the question of whether the shell is formed as an overflow of fluid 

 secretion on to the surface or as a secretion into a cavity vmder ectoderm cells, whether 

 as a secretion from cells, a cuticularization of the exposed ends of cells or a transform- 

 ation of cells themselves. There is a mass of detail yet to be made out with regard to 

 the exact process, the formation of the hinge as distinct from the shell, the origin of the 

 calcareous deposit, the position and serial relationships of the glands that later secrete 

 the different layers of the shell. 



My own observations were at first too much taken up with obtaining 

 a balanced perspective of the various stages of development to permit 

 leisure for such special problems — ^indeed I was not aware at the time of 

 this particular case. It must be understood that the period of oyster 

 propagation is a somewhat brief period for each year, and that there are 



