OF TUb; TRlLOlUTliS. 29 



Asaphiis c.vpa/isia s. conii/jeruii belongs to the form with iiulistiuct rings in the axis of the 

 caudal shield ; I have counted in it six siiort articulations, and a long, oval, terminal articula- 

 tion. In another imperfect one, I believe I recognize nine rings, and a shorter, almost 

 circular, terminal ax'ticulation. In .-Isap/ius ti/ra/iiiiis, of which I know only the caudal shield, 

 represented in Table V, Fig. 4, there are nine articulations, together with a long oval arti- 

 culation. Indeed there seems to be an articulated axis, without elevated lateral ribs of the 

 shield in most species of Asapkm, while the species referred to the group Isoteles might also 

 belong to this, their ai'ticulation being merely very slight. I have seen no species of Amplw:^ 

 without articulation at the axis. 



On the other hand, we miss the articulation entirely in Hheni/.s and Broutcs, so that both 

 these genera are representatives of the third form of the caudal shield. 



The caudal shield corresponds almost completely in point of size and form with the 

 cephalic shield in the genera AsapJms, Illanus, Ampi/x, and Trinudeus or CryptolUlms -, it is 

 smaller in all other genera, because some of the body rings belonging to it in those genera 

 have become isolated independent rings. Its size therefore decreases with the number of 

 rings, and becomes smaller in the species of Phacops, Cal^mene, Paradoxides, Conocephalm, 

 Ellipsocophnhm, and 0/eji//s, in the latter consisting only of one or two rings, ^onia or 

 Gerastos, a genus which we have already mentioned as the t3'pe of a peculiar structure, 

 is at the head of this series. A certain limit therefore seems to have been placed to the 

 number of the body rings, and those of the abdomen seem to increase when those of the 

 thorax decrease. Emmerich, indeed, has already considered this as the correct relation, 

 but a more particular investigation does not confirm this view ; and, indeed, the fallacy 

 of such a conjecture may be proved by the mere comparison of the species of Phacops 

 among one anothei', inasmuch as they never possess more than eleven thoracic rings, and 

 yet fluctuate between nine and twenty-one in the number of their al)dominal rings. The 

 same thing is also seen in Calymene, but the limits of the series are not there so very dif- 

 ferent from one another, but merely fluctuate between seven and eleven. {Cal. polyioma, 

 according to Dalman.) It appears, however, that the rings of the thorax and abdomen 

 together do not generally exceed the number of thirty, and that in many Trilobites the 

 number in both divisions of the body does not amount to so man)% while the total number 

 of rings is quite uncertain where the articulation at the abdomen cannot be recognized. For 

 the rest, I have only to observe that the divisions at the axis of the head of Trilobites 

 are likewise nothing more than indications of rings, but this can be easily reconciled with the 

 view I have before expressed, namely, that they may be looked on as protuberances of the 

 gill muscles situated beneath them, since as many body rings are always missed in all 

 Crustacea as there are accessory pairs of gills at the head ; from which it is evident that 

 every pair of gills is aflixed to a particular ring, the latter, however, losing its independency 

 by its intimate junction with the head. Since also the number of lateral furrows of the head 

 is never more than three, by which, however, there are never formed more than four pro- 

 tuberances, we might assume as many gills in the Trilobites, and suppose that in all cases 

 where these protuberances are wanting, and where the anterior lobe contains all the others 

 within itself, one pair of the gills must have grown very large (this would be the first pair 

 according to analogy), whilst the others have disappeared, although they have not perhaps 



