1878.] MR. R. COLLETT ON CERTAIN GOBIOID FISHES. 327 



The mucous glands' 1 , the true nature of which has perhaps not 

 yet positively heen proved, are to be found on all parts of the head 

 in longer or shorter rows, which vary comparatively little in the 

 different specimens. To give a more detailed description of them 

 would require too much space ; but a drawing (see fig. 2, p. 323) has 

 been taken from a comparison of several specimens. Two parallel 

 series across the hinder part of the interorbital space are easy to be 

 observed, as may also two others, running along the upper side of the 

 snout, and sending forth some shorter series down to the nostrils. 

 The rest of these glands reach as far back as to the first dorsal fin ; 

 and shorter rows are found immediately behind the veutrals. 



The scales and the skin. — The whole body, with the exception of 

 the head and fins, as well as the neck, throat, and belly, down to the 

 vent, is covered with large and very easily deciduous scales. These are 

 fastened to transverse muscular impressions, which bend together in 

 the lateral line of the body, where the vertebrae are connected. The 

 scales are arranged along the sides of the body thus: — First a 

 series of 25 scales are fixed along the lateral line ; above and 

 below this row run two other rows ; so that together there are 5 rows 

 between the dorsal and ventral line. The scales are perfectly trans- 

 parent, membranous and cycloid ; in the lateral line they are a trifle 

 smaller than in the other rows ; they are comparatively large, having 

 a diameter of nearly 1 millim., and are few in number. Accord- 

 ing to the above statement the total number of scales would be hardly 

 more than 100 on each side of the body. The period in which the 

 scales are developed I have not been able to observe to a certainty, as 

 the scales in young specimens seem to be still more easily deciduous 

 than in the older. In the smallest specimen I have found, with 

 a total length of 13 millims., the transverse impressions, to which the 

 scales are fastened, are already visible ; but I have not detected the 

 scales themselves before the specimens have attained about half their 

 size. 



The colour. — When alive the whole body is almost perfectly trans- 

 parent, having only a few minute stellulate pricks in the skin. One 

 may without difficulty count the vertebrae from the outside, and 

 clearly distinguish the brain with its different parts, as well as the 

 pulsation and the blood-corpuscles. The most coloured part of the 

 body is the eye, and the innermost part of the black bladder. When 

 preserved in spirits, or in a frozen state, they soon become white and 

 opaque, whilst if preserved in glycerine they remain half transparent ; 

 then the minute pricks become fainter, whilst the iris always retains 

 its silvery colour. 



In the breeding-season the minute pricks become more numerous, 

 especially in the males. A row of these pricks, yellowish red in 

 colour, stretches along the sides, corresponding in number to the 

 vertebrae. Along the dorsal line another row of brown pricks is 

 found. Along the anal fin, on each side of the root of the rays, is a 

 row of black pricks, which continues behind the end of the fin in 



1 " Stripes of minute warts," Giinther, Catal. vol. iii. 1861 ; " External 

 papilla of taste," Winther, Naturh. Tidsskr.3. R., 9. B., p. 181. Kbhvn. 1874, 



