THE FAUNA OF THE ST. JOHN GROUP. 107 



more tumid cheeks than the type, or var. a., and the fulness extends well down toward 

 the genal angle ; the furrows are generally deeper and more abrupt, and the anterior mar- 

 ginal fold usixally broader and flatter. The occipital spine is long and stout, and the 

 ociilar ridge frec[uently not observable. The largest head of this variety found was 15 X 

 30 mm. The marked feature of this variety is the profusion of spines that cover the raised 

 parts of the cephalic shield, — usually double the number found on the typical form. 



Var. y. perhispid/is {Fi'g. 12), is distinguished from the others in having more than 

 one row of tubercles or spines on the lateral limbs of the anterior margin ; and by the 

 more numerous and smaller tubercles, almost uniform in size, that cover the test. The 

 largest test of this variety found was 15 X 30 mm. 



GrRClWTH AND DEVELOPMENT OF THE YoUNG. 



Fortunately in this species the peculiar form of the cephalic shield, and the hispid 

 surface enable us to recognize without difficulty very small tests ; in the very youngest 

 we lose the guidance of spines, but the general form is a sufficient assurance that, in the 

 minute tests to which I am about to refer, we have the embryonic and larval stages of 

 this trilobite. The youngest form, however, differs widely from the adult, and, without 

 the intermediate links, one woiild hesitate to assign them to the same species or even the 

 same genus. In the following descriptions I have named as " stages " those tests, where a 

 new featiTre is introduced in the moulding of the cephalic shield during growth. 



Embryonic Stage (?), 1 X IJ mm. (Figs. 14 and 15.) — Shield semi-oval in outline 

 and globose, with genal spines. The anterior margin is bounded by a very thin, thread- 

 like fold at the anterior c[uarter. The genal spines are slender, arched inward at the 

 points, and about as long as the cheeks are wide. 



The axial lobe of the cephalic shield (there being as yet no glabella) is club-shaped, 

 the anterior half being enlarged ; it is bounded throughout by distinct furrows. 



The occipilal ring is not distinguishable as a separate part, bu.t the occipital spine 

 appears as a distinct protuberance at the posterior end of the median lobe. 



The posterior viargin is strongly arched forward on each side of the centre, and back- 

 ward again on nearing the genal spines ; the fold is visible, but is a mere thread. 



The cheeks are somewhat tumid oiitward, but they fall below the level of the median 

 lobe in the forward half In a few tests of this size the ocular ridge is faintly outlined for 

 a short space on the anterior slope of the cheek, on each side of the axial lobe and c[uite 

 close to the anterior marginal fold, but usually it cannot be distinguished. 



I have marked this form with doubt as an embryonic stage of Ct. Mattheici. It is not 

 found in such groat numbers as the succeeding stage, and may perhaps be only an 

 immature condition of it ; or possibly it may be an earlier stage of one of the other Cono- 

 corypheans. The club-like form of the enlargement of the front of the axial lobe may be 

 clue to backward pressure upon this very flexible test. In most of the tests of the next 

 stage obtained from the slates, the width of the anterior end of the axial lobe is, on the 

 contrary, exaggerated by downward pressure. 



First Stage, IJ X 2 mm. [Fig. 16.) — Shield semi-circular, with genal spines. 

 Anterior margin. — The fold faint, thin and thread-like, and extending about one-third 



