FREE INSTITUTE OF SCIENCE 



863 



TERTIARY FAUNA OF FLORIDA 



it the scar of the Hgament. The ligament and cartilage start in the youngest 

 stage of the shell from a point immediately under the beaks. Once started, 

 each remains continuous between the two valves of the shell and persistent, 

 so that as they lengthen with growth the portion below each umbo remains 

 intact. According to the amount of expansion of the shell margin between 

 the beaks and the direction of growth of the valves, the form of the adult 

 cartilage and ligament may be crescent-shaped, with a posterior convexity; 

 sagittate, like a barbed arrowhead ; or lanceolate, like a leaf-shaped spearhead. 

 The distance between the points of the barbs is determined dynamically by 

 the distance between the umbones of the valves : when they are widely 

 separated, as in Mactra Spenglcri, we have the most extreme crescent shape; 

 when they are but slightly separated, the sagittate form ensues ; when the um- 

 bones are close together, the species must have a lanceolate ligament. A 

 steep slope of the dorsal shell margin backward from the umbonal region 

 necessitates a short ligament, while a nearly horizontal long posterior cardinal 

 margin promotes a long and narrow ligamentary connection. The correla- 

 tions are purely dynamic. There is little doubt that the existence of a 

 separate resilium and ligament is due to mechanical forces acting on a thick 

 hgament, as I have elsewhere shown.* Why the ligament should become 

 embedded in the cardinal border so as to become subject to these forces is not 

 so clear, but is probably accounted for in part by the fact that the hinge-hne 

 is rigid in proportion to its length, and, in general, if high up in the dorsal 

 arch it must be short, and can gain length only by descending. Whatever 

 the reason may be, it is doubtless analogous to that which would account, in 

 a species where ligament and resilium have become fully differentiated, for the 

 further subsidence of the ligament until it, in its turn, may be wholly sub- 

 merged below the cardinal margin, so that the latter closes over it, leaving no 

 ligamentary substance whatever e.xternal to the shell. In the Mactridce every 

 stage of this process may be observed, from the condition where we have a 

 marginal external ligament, walled off by a lamina of shell from the resiliary 

 pit, to one where ligament and resilium occupy different portions of a single 

 cavity, wholly invisible from the exterior when the valves are closed. 



The shelly portions of the hinge arise from a shelly basis stretched 

 antero-posteriorly between the limbs of the arch forming the cardinal margin. 

 This basis is called the hinge-plate, and it may have its surface "flat" 



* Am. Journ. Sci. and Arts, vol. xxxviii., Art. 55, Dec, 1889. 



