of the Fishery Board for Scotland. 



265 



is similar to the first, but the expanded base is much shorter, and forms a 

 slightly forward projecting process. Posterior to these, and as far as the 

 24th vertebra, the neurapophyses all have a similar form — a rounded 

 articulating head expanding into a somewhat lozenge-shaped portion (when 

 looked at in a backward and lateral direction) the lower angle of which 

 passes into the head ; the upper is continued upwards as the spine, and 

 the anterior median angle forms a projecting sharp process shutting in the 

 neural cord laterally, while the outer median angle passes backwards as 

 the lateral appendage of the neurapophysis up to the 23rd vertebra, where 

 a change occurs to be described later. Above this again, and forming a 

 roof to a secondary canal for the longitudinal ligamentous band, is the 

 sharp process above referred to, projecting at first forwards and somewhat 

 inwards, but further back forming a bar across and projecting slightly in 

 front of and behind each neurapophysis, the pair being closely approxi- 

 mated in the middle line. This disappears where the neurapophyses become 

 fused at their upper end (PI. XVI. fig. 6). Although the vertebral zygapo- 

 physes are present from the first, they do not overlap the neurapophyses 

 until about the 23rd vertebra, where the anterior zygapophyses commence 

 to be formed on the neural arches. Here the basal expansion already 

 described begins to disappear; its lower angle formed by the anterior edge 

 and the base of the articulating head (and which is not visible as more than 

 that anterior to this point), now enlarging continually backwards and 

 forming the sharp pointed dorsal anterior zygapophyses, which reach their 

 maximum length about the 50th vertebra. 



As in the hsemal arch the posterior (vertebral) zygapophyses disappear at 

 the 52nd vertebra, and the neurapophyses come to slope so much back- 

 wards as to touch and fuse with the posterior edge of the vertebra on which 

 they arise, and they even here form an articulation with the succeeding 

 vertebra, the anterior edge of which fits into a small notch in the lower 

 (posterior) surface of the neurapophysis, which lies close down on it. 



The Intermusculars. 



Along each side of the body, ventral to the lateral line — which almost 

 exactly corresponds to the line of vertebral column — are arranged a series 

 of delicate bones — intermuscular bones — situated in the connective tissue 

 between the myotomes (PI. XV. fig. 1, i.m.; PI. XVI. figs. 4, 5, i.m.). These 

 bones do not articulate directly with any other part of the skeleton. They 

 are developed in the intermyotomic fascia, and are, in the anterior half of 

 the herring, attached at their proximal ends by a short (2 to 3 mm.) liga- 

 ment to the ribs near their heads (in 292 mm. fish about 3 to 4 mm. 

 from the head). From this point they pass, with a slight curve, outwards, 

 backwards, and downwards, lying (owing to the slope of the sides of the 

 masses of muscle) with a more pronounced inclination backwards, and a 

 greater curvature outwards, than do the ribs. The distal portion of each 

 intermuscular bone in this part of the body lies therefore almost directly 

 external to (and about 2 mm. from) the rib following that near the head 

 of which it arises. The only departure from this arrangement in the 

 anterior half of the body is that where the ribs, as already described, are 

 free, i.e., not closely applied to the vertebras, the short ligament fixed to 

 the inner end of the intermuscular arises not from the rib itself, but as a 

 branch of the thread of connective tissue attaching the rib head to the 

 haemal arch, and in such a way as to pull principally on the latter. 

 There is a short intermuscular in front of the 1st rib, i.e., the first inter- 

 muscular corresponds in position to the 2nd vertebra. 



