﻿56 18 



Amphisile. Fig. 5 is of A. sciitata in which tlie mova'ble spinous ray is laciving and 

 the muscles belonging to it being superfluous are atrophied; but, in the main 

 features, even down to details, the cross-section agrees with that of strigata, except 

 that in the latter the line of division between the two fused interspinous bones is 

 much more distinct; in several cross-sections it is only seen right in the middle 

 between the two upper canals, sometimes also on the one or the other side; it is 

 thus made somewhat more distinct in the figure than it usually appears in this 



species. This figure also shows 

 — : 3 that the bony mass of the dor- 



sal spine has quite a complicated 

 structure which I have not suc- 

 ;- 3=»— — ^- ceeded in fully clearing up; it 



' ... , is clear, however, that the peri- 



wig. 7. Î > 1 



A,upi,isih' siriaaia. End oi tin- dorsM spin, wiih spinous n.y «, pheral portion of the bony mass 



hitciMl bony piece ; and spinous piece /. The dotted lines indicate Jogg j-,Qt belong to the illterspi- 

 llie position of the muscular tendons. , i i • 



nous bones themselves but arises 

 from the exoskeleton; this is shown, amongst other things, by the fact that the 

 canal for the lateral line (see fig. 5) is enclosed in the dorsal spine. The statement 

 made above that "the longitudinal keel of the interspinous bones broadens out in 

 the skin" must therefore be taken with some reservation. 



Considering now the upper end of the dorsal spine and its connection with 



the movable spinous ray in A. strigata (fig. 7) and punctulata, we find that there is 



no ray to the interspinous bone 3; the spinous ray is attached to 4. The lower 



end of the spine is cleft and sits on the upper, wedge-shaped 



P end of this interspinal (fig. 8); on each of the lateral surfaces 



of the wedge we find a series of 5—6 concentric, half-circular, 



very sharply marked ridges (figs. 8, 9); into these fit very 



^'^'^■.,. . , exactlv corresponding ridges on the inner surface of the cleft 



Aniph. singcita. Hind end •' r o o 



oftiorsni spine, seen from gf ^j^g spinous rav. This arrangement excludes all other kinds 



below ; spinous ray and f J o 



removed •''<i^™pe'nin"'*'for ^^ movement than ouc in a vertical plane, but on the other 

 priLor oTlfpinnus îay'fl hand makes the joint unusually firm and steady; the joint is 

 further supported by the small plate I (fig. 7) which covers 

 both sides of the end of the interspinous bone and a part of the base of the ray. 

 On the side of this plate which covers the ray we find some slightly raised, circular 

 ridges and others corresponding to these are seen on the outer surface of the base 

 of the ray. The angle through which the ray can move is not very large; in an 

 upward direction the ray can only turn until it is almost in line with the dorsal 

 spine, downwards so far that it lies almost parallel to the small spine / (fig. 7). It 

 is probable that the fish can at will fix the spinous ray immovably at any point 

 between these limits; the joint is indeed of the same type as the corresponding 

 articulation in Centriscus, where W. Sorensen (32) has distinguished it as a "stop- 

 joint". 



