1892.] WILLIAMS — HIXGED BRACHIOPODA II7 



coils, calcified in the extinct Spirifericlc-e, were the supports of fleshy 

 brachiae, as in the living Terebratulidoe, and the question which arose 

 was how to explain the difference in direction of coiling of the arms 

 in the two types. The solution of this problem resulted in explain- 

 ing the relationship of the two families. 



The difficulty was so great, that I made some models of the 

 lamellae of thin brass and made spiral coils of copper wire and 

 experimented with the models to see how one could be changed into 

 the other. 



With this device, imitating the characters of the loop of the 

 living Waldheima and its flesh}^ spiral arms, I discovered that doub- 

 ling back the lamellae with the spirals attached beyond the bend 

 caused a reversal of the direction of the coil from that presented by 

 it when attached before the bend, thus producing the exact difference 

 observed on comparing Waldheima and Anazyga (see figs. 2 and 6.) 



This experiment showed that the fundamental difference between 

 the brachial apparatus of the two families does not consist alone in the 

 presence of a calcified spiral in one and its absence in the other, but 

 in the fact that in the Spiriferidte the primary lamellce are continued 

 directly into the spiral coils, whereas in the Terbratulidse the primary 

 lamella on each side is doubled back upon itself to near the position 

 of the mouth from which point the spiral part of the arm begins, the 

 reversal of direction of the coils of the spiral resulting from this 

 reflexion of the primary lamellae. 



This difference and the relationship is best seen in the earlier 

 types, as Anazyga (fig. 6.) (See also Dayia, fig. 7), and Waldheiinia 

 (figs. 2 and 4). In both of these forms the primary lamellse (/) are 

 long, but in Anazyga and Dayia the bridge [b] connecting the lamellae 

 of the two sides sets off before the beginning of the spiral coil, 

 while in Waldheimia (4) the bridge, or connecting part {b of fig. 2), 

 does not occur till after the reflection of the lamellae to form the loop. 



In the short-looped Terebratulidae ( Tei-ebratulina, etc.,) the primary 

 lamellae are not calcified down to the front and the reflected and looped 

 part of the apparatus is not supported by calcified lamella. In the 

 greater number of the genera of spiral-bearing brachiopoda (the Heli- 

 copegmata) the bridge connecting the lamellte is higher up than in 

 Anazyga, Zygospera and Hindella. Thus pointing to the conclusion 

 that the earlier forms were all with long primary lamellae, and that 

 the forms in which the primary lamelLx were so far extended forward 

 as to be reflected backward before the coiling of the arms, developed 



