(JAid'IKE. 

 Here, as in tlie Mackerels, the pointed sliajic (if'tlie 



549 



tins indicalcs a cajiacity of s\vif"t indlinn ;iii(l suddrn 



c'liunyes \n Its (hrcctuin. 



pccloi'al tins arc set hi;!li, 



almost vcrtic!ill\ , and (il)lii|iicl\ pointed. Tlicir leii^''th, 

 which at the end of the sliacll\- invcidh' ehaii;;'os of 

 growth is ahout e(|ual to the 1ciil;i1i <it' the postorbital 

 part of the head, snl)sc{|nentl\- shows i-elati\c diminution, 

 but still measures aliout '.)() '-i, (sometimes 80 %) thereof. 

 'I'iie distance between the front end of tlie insertion of 

 these tins and that of the ventral tins is I'rom about 5 to 

 SVa times the lengtli of the former. They contain 13 rays, 

 the uppermost of whicli is the longest or equal in length 

 to the second va\-, undi\i(h'd and compressed like a 

 sabre. This is also true of tiie tlrst i-ay in the \entral 

 tins, whicli are also pointed". The length of these tins is 

 about ecpial to the height of tiic dorsal tin. They are 

 moved forward dui-ing growth, the distance betw'eeu 

 them and the lijt oi' the lower jaw decreasing in pro- 

 ])ortion to the length of tlie body, while on the other 

 hand, the distance between them and the beginning of 

 the anal fin increases. At the end of the true juvenile 

 changes of growth the former distance is nearlj' ^ .5 

 of the length of the body, but in specimens 7 or 8 dm. 

 lonji- only aliout GO % thereof. The latter distance si- 

 midtancously increases from about 11 % to 14 or lb?6 

 of the length of the body. The dorsal and anal tins 

 are almost alike in shape; but the former is always 

 somewhat shorter and lower than the latter, and the 

 distance between its begirming and the vertical line from 

 the beginning of the anal tin is generally about equal 

 to that between its termination and the vertical line 

 from the end of the latter. The distance between the 

 dorsal fin and the tip of the lower jaw varies, after the 

 end of the juvenile changes, between 79 and 7.') ?o of 

 the length of the body. The distance between its end 

 and the beginning of the caudal tin is somewhat greater 

 than the length of the pectoral tins, varying between 

 about 64 and about 52 % of the length of the tin itself. 

 This is due to the fact that the length of its base un- 

 (lersoes even relative increase from about 1 1 % to about 

 IS % of that of the body. The first two rays are un- 

 divided, the second being about twice as long as the 

 first, and the longest in the whole fin. The pointed 

 lobe at the beginning of the fin, the height of which 



is generally e(|ual to the length of the ventral fins, is 

 formed bv the first six rays; the other rays are gener- 

 all\ ol' miiroiiii lieight, though a slight elongation may 

 be ol)served in tlie middle ones. This is never so mark- 

 ed, however, as the posterior elexation of the; dorsal 

 tin in tlie .Mediterranean Hamphistoma imperiale (Caii- 

 tifiinii). On accoimt of the changes of growtii in the 

 dorsal fin the begiiniing of the anal fin may .sometimes lie 

 vertical h- below that of the dorsal fin, as in our largest 

 specimen; but the base of the anal fin is always longer 

 than that (j1 the dorsal, varying between about \2\ .,% inul 

 about 14 % of the length of the body, and the distance 

 between it and the caudal tin is less than the length 

 of the peetoi'al fins. In adult spceimeiis the caudal fin 

 is dee|il\ forked, with pointed lobes, the length of which 

 is about ' ,„ OI' '/,j of that of the body. The extreme 

 end of each lobe is formed by the outermost branched 

 ray, which in full-grown specimens, however, is hardly 

 any longer than the longest (hindmost) supporting rav. 

 We have always found the number of branched rays 

 in the caudal fin to be 13\ The vent, which is situated 

 just in front of the anal fin, is noticeable for its some- 

 what considerable size. 



The scales. of the body are very thin and for the 

 most ]>art deciduous; but on the back and head they 

 are hard to detach, and extend forward over the nasal 

 region and the base of the upper surface of the inter- 

 maxillary bones. Even on the prcorbital bones small 

 scales may be found, and on the cheeks and opercula 

 they are distinct. On these parts of the body, however, 

 all the scales are small; but on the ventral sides, in- 

 terspersed among the small scales, we find larger ones, 

 almost thrice the size of the former and serving as a kind 

 of foundation for them. The vertical diameter of these 

 large scales is greater than the longitudinal, sometimes 

 twice as great, and sometimes measures as much as Iialf 

 the longitudinal diameter of the eye. Their structure 

 is the same, hoAvever, as that of the small scales: they 

 are thin and transparent, with distinct, concentric rings 

 formed during groAvth, but without radiating stria' or 

 cur\ es. The scales in the rows which coa.st the lateral 

 line are somewhat thicker and curved, while those of the 

 lateral line itself are pierced in the usual manner. Where 

 this line runs on each side along the belly, ri.sing slightly 



" Among tlie live branclied rays in these fins the inner ones, in pnrlinilnr, liavc .1 well-developed derinal fold between them; and as n 

 similar fold also follows the inner side of the innermost ray, we may easily fall into the error sometimes committed, of stating the number 

 of these rays at 6 instead of 5. 



'' According to Lilueborq the number of these rays is 14, according to Day 1.3 or 14, and according to Kroveh 17. Ekstbom gives 13. 



