Yol. 50.] SYSTEMATIC POSITION OF THE TRILOBITES. 419 



It is further of especial interest to note that the lines of fusion 

 between the lateral projections of the first segment and the pleurae 

 of the second segment apparently correspond with the posterior 

 halves of the mysterious cephalic sutures. Many trilobites have 

 these sutures running out laterally, as if dividing the shield into 

 two somewhat similar pleurae (e. g. Cromus intercostatus and Dal- 

 manites). 1 The symmetry of the line itself has, however, been 

 broken by the wandering backwards of the eye-tubercle, which, as 

 we shall see, belonged originally to the first segment, and wandered 

 only secondarily on to its . lateral projections. The larval forms of 

 Olenellus (fig. 3, p. 415) show how this line might run almost 

 straight backwards when the first pair of projections are very 

 largely developed in comparison with the pleurae of the following 

 segments, which, like the pleurae of the rudimentary tail-segments, 

 may slope backwards. 



The retention of the line of fusion 2 between the anterior edges of 

 the pleurae of the second head-segment with the lateral projections 

 of the first segment, as a line of weakness through the thick dorsal 

 head-shield, may have been useful for ecdysis. The thin ventral 

 membrane would no doubt have split easily ; but, for the drawing out 

 of the limbs, etc., it is necessary to open up the dorsal surface. 

 This would have been extremely difficult in the case of the trilo- 

 bites, unless special provision had been made for it. Both Limiting 

 and Apus are said to moult by splitting along the frontal edge. 

 In the trilobites, the splitting generally appears to have left the 

 frontal edge on each side of the glabella and to have run back to the 

 eyes ; it then followed the line along the inner posterior edges of the 

 eyes, which, as above stated, may well have been the original line of 

 fusion of the first and second pairs of pleurae forming the head-shield. 



IV. In endeavouring to deduce Apus from a carnivorous annelid, 

 by the bending round of the first segment, I had assumed that the 

 eyes were originally on the prostomium (as they are typically in 

 carnivorous annelids), and that when this was bent round ventrally 

 they wandered up on to the dorsal surface of the first segment. 

 Clear traces of this wandering of the eyes from the ventral on to 

 the dorsal surface can still be found in the development of Api<?, 

 the eyes showing a gradual dorsal displacement during development. 

 I brought forward also some morphological evidence in favour of 

 this dorsal wandering of the eyes of Apus ; for instance, the position 

 and shape of the brain and antennal nerves seem best explained on 

 the assumption that the brain had been dragged out of its original 



1 An almost similar suggestion was made by M'Coy, 'On the Classification 

 of some British Fossil Crustacea,' Ann. & Mag. Nat. Hist. ser. 2. vol. iv. 1849, 

 who concluded, from the position of the eyes as belongiugto the first 'ring,' 

 that the suture running posterior to tliem was the line of junction of the first 

 and second rings. He claimed the whole sutures as such. I would, however, 

 only claim the posterior portions of the suture, believing that the anterior lobe 

 of the glabella certainly belongs to the first segment. 



2 S. W. Ford, Am. Journ. Sci. ser. 3, vol. xiii. (1877) p. 267, if I understand 

 him aright, states that this fusion is incomplete in the youngest stages. 



