1890.] SPECIES OF FISHES FROM MADEIRA. 453 



of the head, its posterior extremity uot reaching quite so far as the 

 vertical from the middle of the e}'e. 



In the upper jaw there is a narrow band of minute curved teeth 

 with two larger conical teeth ; at the front of the lower jaw a simi- 

 lar band of teeth with four conical ones, and at the sides behind a 

 single series of teeth. Minute teeth on the vomer and palatines ; 

 none on the tongue. The opercular pieces are clothed with scales. 

 There is a spine near the upper angle of the opercle and another 

 spine a little lower down. Lower still are a few serratures. The 

 vertical border of the preopercle is strongly serrate, and there is a 

 spine at the angle with a smaller spine a little beyond. The rest 

 of the lower edge is entire. ^Ul the spines are directed back- 

 wards. 



The dorsal fin begins over the root of the pectorals and extends 

 a little beyond the end of the anal. The first and second spines are 

 short, the third and fourth are equal iu length and have skinny 

 tags at their tops. The soft portion of the fin is higher than the 

 spinous portion ; the last ra}s reach to the base of the caudal. 

 The jiectorals are not quite so long as the ventrals, and they reach 

 back to the beginning of the anal, or to the end of the spinous por- 

 tion of the dorsal ; their bases are scaly. The ventrals are inserted 

 under the root of the pectorals and reach back a little beyond the 

 commencement of the anal. The second soft ray is elongate and 

 filiform. The anal begins a little behind the middle of the base 

 of the dorsal. The first spine is short, the second stout and the 

 longest of the three. The soft portion of the fin is higher than the 

 second spine, but the last rays do not reach nearly so far as the 

 base of the caudal. The caudal is deeply furcate and has some of 

 the exterior rays elongate and filiform. The membrane between 

 the rays is scaly as in Callanthias. 



The lateral line has 37 scales ; it rises from the edge of the gill- 

 cover, runs high up on the side following the curve of the back, 

 descends rapidly under the end of the base of the dorsal, and passes 

 along the middle of the tail to the base of the caudal. 



The colour of all the specimens had faded to a pale brown ; but 

 faint traces of pink or red were visible in parts, and there seemed 

 to have been twelve or more narrow transverse bands on the sides, 

 alternating with paler bands. 



This fish much confuses the distinction between the genera An- 

 tJiias and Callanthias. With the only known member of the latter 

 genus it agrees in having only six brauchiostegal rays, in possessing 

 much the same external form, and in having the angles of the caudal 

 fin prolonged into filaments. It would therefore appear that the 

 difference between the two genera is reduced to the serrature of the 

 border of the preopercle, which is present in Anthias and not in 

 Callanthias, a trifling artificial distinction. 



The first specimen that came under my notice was found dead and 



dry by myself, in a cavity amongst a mass of Ostrea and Chama 



shells brought up from deep water. Afterwards four specimens were 



taken by the officers of the ' Britannia ' whilst engaged in repairing 



Proc. Zool. Soc— 1890, No. XXXI. 31 



