40 THE AMERICAN NATURALIST [Vol. XLIIl 



although far from being as satisfactory as fresh, living 

 material, shows oyster larvae (Figs. 1, 6) but little older 

 than the stages at which the observations of Brooks 

 closed, viz., six days old from the date of fertilization. 

 Brooks did not give measurements, so that it is impossible 

 to be exact on this point— I can only judge from the shape 

 and organization. Plankton collected July 11, 1904, be- 

 tween Curtain Island and Ram Island contains an 

 abundance of minute, transparent bivalve larva? (phylem- 

 bryos) in what may be known as the straight-hinge stage 

 to distinguish them from the older larva 1 with high umbos 

 (the umbo stage) that obscure and modify the hinge line. 

 A hasty and superficial observation of these combined 

 with the fact of their occurrence in proximity to oyster 

 beds might easily lead to the conclusion that they are all 

 oyster larvae. But they are not. Many of them are 

 clams, a few are mussels, and one in a great number is 

 an oyster. A full statement of how I have determined 

 this would require too great a digression and will be 

 dealt with in another paper, but it results from a com- 

 parative study of bivalve larva? in the different localities 

 of the Biological Station combined with researches into 

 the distribution of adult forms. Adults of the above- 

 mentioned genera are easily distinguished ; the full-grown 

 larvae less easily, for, since they bear little resemblance 

 to the corresponding adults, other marks of distinction 

 have to be selected; but the young larva? are still more 

 difficult, for, according to the biogenetic law, the younger 

 they are the more nearly they resemble some stage of the 

 common original ancestor and of course approach one 

 another in likeness. Under such conditions the prac- 

 ticable, distinguishable characters may again be different 

 and require a more critical scrutiny. Since I first turned 

 my attention to bivalve larvae I have found it necessary 

 to change my point of view and mode of procedure. One 

 can not safely trust to the eye in judging proportions, 

 but must resort to a definite and unvarying method of 

 measuring by means of ocular and stage micrometers. 



