April 10, 1914] 



SCIENCE 



543 



Douville, like Jackson, considers the oysters as 

 direct descendants of true aviculoids not from 

 Perna, however, but from Lima. The sessile 

 aviculoids are cemented, as a rule, by the 

 right valve, while in the oysters it is the left 

 valve that is attached. However, it is probable 

 that Douville overvalues the systematic im- 

 portance of this character, for in the inequi- 

 valved Pectens it is sometimes the right valve 

 and sometimes the left which is the more con- 

 vex. Lima, a significant fact to Douville, 

 assumes an almost vertical position as 

 indicated by the approximately equal valves 

 and the presence in both of a byssal opening. 

 Furthermore, the characters of the ligament 

 are similar in the Ostreas and the Limas. The 

 Ostrea, however, does not apparently attach 

 itself by a byssus at any stage in its develop- 

 ment. Jackson watched very carefully for 

 such a phenomenon, but was unable to find any 

 trace of it. On the contrary, the spat appear 

 to attach themselves at the very beginning of 

 cementation by the margin of the reflected 

 mantle. If the Ostreids were the direct de- 

 scendants of Lim,a, as Douville believes, 

 they would probably reveal to a skilled embry- 

 ologist such as Jackson some clue to the pres- 

 ence of a former attachment. 



Two major groups of Ostreids have been 

 established by Douville: in the one, he has 

 assembled all the dominantly smooth forms, in 

 the other, all the dominantly plicate. In both 

 groups, he finds forms with straight umbones 

 and those with gyrate beaks, and the outline of 

 the species is, in his opinion, directly asso- 

 ciated with the environment. Thus, the 

 straight-beaked Pycnodonta is characteristic 

 of deep and quiet waters, while forms with 

 strongly twisted umbones, such as Exogyra, 

 are developed in the more shallow waters, 

 where there are strong currents to be resisted. 

 If this theory be accepted, it is difiicult to 

 account for the frequent association of Pycno- 

 donta and Exogyra in considerable numbers 

 in the same marl bank. In his correlation also 

 of the sculpture of the plicate group with 

 their environment, his theories seem unwar- 

 ranted by the facts. It is true, to be sure, 

 that, as a rule, the right valve of the Ostreids 



is less vigorously sculptured than the left, but 

 there is no evidence that the sculpture evan- 

 esces more rapidly in the littoral forms. On 

 the contrary, a strong ribbing is most fre- 

 quently developed where the need for resist- 

 ance is greatest. 



Douville as well as Jackson considers the 

 Gryphceas, Pycnodontas and Exogyras as de- 

 rivatives of the true oysters. Jaworski does 

 not admit, however, that the Ostreids of the 

 Triassic which have served as the theoretical 

 ancestral types are true oysters. Steinmann 

 traces the ancestors of the group even back of 

 the Mesozoic and considers Eurydesma of the 

 Permo-Carboniferous of Australia and India 

 as the true ancestor of the Ostreid stock. This 

 form is characterized by a heavy, lamellar 

 shell, prosogyrate beaks, a marginal, poster- 

 iorly produced ligament, and possibly an in- 

 cipient dentition. The most significant fea- 

 ture, namely, the character of the muscle im- 

 pressions, is doubtful : the form has been 

 commonly accepted until recently as mono- 

 myarian. Morris, however, observes " there is 

 one large impression posteriorly and perhaps a 

 small one anteriorly." If the presence of an 

 anterior as well as a posterior scar can be 

 established, Eurydesma would, in the opinion 

 of Jaworski, fall in line behind Heteros- 

 trea; if its absence can be proved, Eurydesma 

 should be considered as the ancestral form of 

 the Gryphwas and Exogyras. It is Heterostrea 

 steinmanni Jaworski which its describer con- 

 siders the true ancestor of the true oyster. 



The line of development between the two 

 forms is, in his opinion, well defined, and easy 

 to follow : the degree of coiling of the umbones 

 is functional upon the ratio of the size of the 

 left valve to the adhering surface; the larger 

 the area of cementation, the stronger the 

 tendency toward an elongate outline and 

 straightened umbones; with the change in the 

 direction of the beaks, there is a corresponding 

 shift of the ligament from its original position 

 along the posterior margin to a more efFective 

 point of attachment directly beneath the tips 

 of the umbones. The obsolescence of the 

 anterior adductor is doubtless the result of the 

 shift of the antero-posterior axis consequent 



