MULLIDAE OF THE WESTERN NORTH ATLANTIC 



441 



h 



^ 



Figure 31. — Cleared and stained palato-vonierine tooth 

 patch of 37.0-mm. Altillus auratiis. Line equals 0.5 mm. 



Scales 



Lateral line scales were counted on 28 specimens 

 ranging in size from 44.8 to 199.0 mm. The range 

 in counts held for small sizes through the largest 

 specimens, 29 to 35 (mean of 32), with 32 and 33 

 most frequently encountered. This is somewhat 

 lower than the count of "about 40'' given by Jor- 

 dan and Evermann (1896) and Beebe and Tee- 

 Van (1933) iov M . auratus. 



The extreme deciduousness of the scales during 

 the pelagic stage is demonstrated by the following 

 figures. One hundred and eight M. auratus, rang- 

 ing in size from 9.3 to 36.4 mm., were examined 

 in an attempt to determine the manner in which 

 scale formation occurs. Fifty-one of these speci- 

 mens, ranging from 9.3 to 34.4 mm., were totally 

 devoid of scales. The 57 specimens with one or 

 more scales ranged from 21.8 to 36.4 mm., and 

 none had more than four scales in the lateral line. 

 Scales present in the specimens below 30 mm. were 

 usually around the dorsal fins and restricted to 

 above the lateral line. 



In a second series of 33 slightly larger speci- 

 mens ranging from 19.6 to 45.8 mm., 12 specimens 

 (19.6 to 37.4 mm.) had no scales. Two in- 

 dividuals, 44.8 and 45.3 mm., had 30 lateral line 

 scales — these were the smallest specimens on which 

 scale counts were possible. 



A srale from a 25.5-mni. specimen lacked ctenii, 

 but one from a 45-nnii. specimen had about 10 

 ctenii. There were 65 ctenii on a scale from a 



117-nmi. specimen, and 186 on a scale from a 199- 

 mm. adult. 



Barbels 



According to Lo Bianco (1907), Jobert (1872) 

 was the first to advance the theory that the barbels 

 of the family Mullidae are nothing more than 

 displaced branchiostegal rays. Zincone (1876) 

 had given the theory some consideration, but was 

 forced to the conclusion that he could neither deny 

 nor confirm the hypothesis, as the question could 

 only be resolved embryologically, and he had 

 Iwen imable to obtain fertile eggs in aquaria. Lo 

 Bianco (1907) quoted Raffaele (1888) as saying 

 that, whereas he (Raffaele) was able to obtain the 

 fertile eggs of Mulhis surmulletus, which hatched 

 in 3 to 4 days, the larvae died in 7 or 8 days after 

 hatching, and there was no trace of tactile barbels 

 at this stage of development. . Lo Bianco said that 

 he too had seen eggs laid in captivitj' by MuUus, 

 and also effected artificial fertilization of both lo- 

 cal species {Mullus surmulletus and MuUus har- 

 bafus), but the larwie died as soon as the yolk-sac 

 was absorbed. 



In the summer of 1906, Lo Bianco obtained a 

 series of Mullus from the Gulf of Naples, 6 to 

 15 mm. in length, plus a few specimens 22, 26, and 

 30 mm. in length. From these he gave a detailed 

 description of the development of the barbels, and 

 in the summation of his work he stated (1907), 

 "From this observation of mine is demonstrated 

 the origin of the barbels of Mullus from the first 

 branchiostegal ...."* 



Also, Montalenti (1937) made this statement in 

 regard to the Mullidae (translation), "In the pe- 

 lagic individual about 35 mm. long the fii-st bran- 

 chiostegal ray detaches from the membrane and 

 constitutes the barbels characteristic of this fish." 

 Tlie series of M. auratus I examined demon- 

 strated the barbels developed from the bone that 

 appeared in the 8.2-nmi. specimen to be the first 

 branchiostegal ray. The barbel development is 

 similar to that described by Lo Bianco for a Medi- 

 terranean Mullus, but occurs at a smaller size. 



Figure 32 shows the position of the branchi- 

 ostegals of an 8.2-mm. M. auratus which agreed in 

 size and shape with those in Lo Bianco 's illustra- 

 tion of an 8-mm. Mediterranean Mullus (1907, 

 fig. 2) . Lo Bianco 's measurements were probably 



* My translatiou of Lo Blanco's statemoiit. 



