150 AMERICAN JOUKMAL 



port. Both the neural and hffini'il lobes secrete those deposits, 

 which in old specimens, as shown bj M. Deslongchamps, often 

 obliterate the mesial ridge and form a solid bed of calcareous 

 matter. Now the loop of other genera is only found upon the 

 hajmal valve, it does not increase in (relative) size with age, nor 

 become abnormally calcified or thickened. 



Again, we find no submarginal loop in the neural valve of 

 other genera, with which the submarginal calcareous deposits 

 which exist in the neural valve of some species of Tliecidium 

 can be homologized. (See Davidson, Mem. Lin. Soc. de Norm. 

 p. 153, pi. viii, fig. 15, T. vermiculare). 



These facts show that no true homology exists between the 

 two formations. Mr. Davidson (in Mem. Lin. Soc. de Norm.) 

 in his classification of the Bracldopoda, hints at this view of the 

 matter, which has been made clear by the further researches of 

 M. Deslongchamps. The resemblance between the lobed struc- 

 ture of T, digitata and the loops of Megathyris decoUata, is 

 due to a similar form of the soft mantle edgt in both, and by no 

 means justifies the deductions of M. Suess in his paper on the 

 internal structure of the Thecidice. 



The spiculas are frequently deposited upon the hinge margin 

 and crura, which has led to further misconception, and induced 

 the belief of the identity of the two dissimilar structures. 



I would add that I consider the gap between this section and 

 the others as far wider than between any two other sections of 

 the group, and that no sufficient reasons have yet been adduced 

 for uniting the family Thecidiidce with the Terehratulidte. Nev- 

 ertheless, for the present, I would defer to the opinions of the 

 learned naturalists previously mentioned. I think sufficient in- 

 formation has hardly been brought forward to justify the opinion 

 of Lacaze-Duthiers that Thecidium in its young state is not 

 pedunculate. Judging bj analogy, it certainly should be ; and 

 if it is not, it would almost be a sufficient reason of itself for 

 separating the group as a separate family. 



M. Suess divides the group into five lesser groups, typified 

 respectively by T. hippocrepis, Goldf., T. papillata, Schloth., 

 T. digitata, Sby., in the cretaceous beds ; T. triangularis, D'Orb., 

 and T. sinuata, Desl., from the Great Oolite and Liassic strata. 



M. Deslongchamps proposes an arrangement of six groups, 

 typified by T. Ferieri, E. Desl., T. maT/alis, E. Desl., T. sinu- 

 ata, E. Desl., T- triangularis, D'Orb., T. pajnllata, Schloth., and 

 T. digitata, Sby. 



In the present state of our knowledge of the subject I am not 

 convinced that any of these groups are sufficiently well marked 

 to constitute subgenera. The only recent species of Thecidium 

 are : 



