NO. 2223. CHITONS FROM THE NORTHWEST PACIFIC— BERRY. 5 



entire series to /. haJcodadensis. Not only was Carpenter's original 

 material from Hakodate, but a considerable degree of variation simi- 

 lar in character to that remarked upon is recognized in the excellent 

 description by Pilsbry, while furthermore the moimted radulae taken 

 at random from the material before me show very little variation. 



Superficially the shell of /. hakodadensis is so generally similar to 

 that of /. alhrecMi that I was not prepared for the striking differences 

 to be found in their radulae. Although seemingly a much less power- 

 ful affair than the strong radula of albrecJiti, that of the present spe- 

 cies is in many respects so much more comphcated that I have not 

 yet been able satisfactorily to elucidate all its details, nor can I secure 

 preparations offering a reasonable coincidence with the sketches of 

 two of the teeth given by Thiele. The latter divergence is possibly 

 explicable on the assumption that the specimens which we have inde- 

 pendently referred to the same species are not really conspecific. 

 Here the major laterals are strongly bicuspid, with the inner cusp 

 conspicuously the larger and longer, and bear a conspicuous wing- 

 like expansion just below the crown but so narrowly attached at its 

 base and hence so easily broken away that it is not always readily 

 observable in dissected radulae and therefore does not show in most 

 of my camera sketches (pi. 3, figs. 4, 5). 



The valves of one of the specimens from Hakodate are sHtted as 

 follows: Anterior valve, 19; intermediate valves, respectively, 2-2, 

 3-3, 2-2, 2-2, 2-3, 2-2; posterior valve, 15 shts; interior bluish or 

 brownish white. A Mororan specimen shows the formula: Anterior 

 valve, 14; intermediate valves, 3-2, 2-3, 3-3, 3-3 (or 4?), 2-2, 2-3; 

 posterior valve, 14 slits; interior deep slate blue. 



The beautiful zigzag sculpture of flattened overlapping pointed 

 pustules, which covers the central areas of young specimens, is 

 usually eroded to the pitted appearance characteristic of the adults. 

 It shoidd be added that the girdle scales of my specimens seen in situ 

 from above under a relatively high magnification are very weakly 

 striate, not smooth, as described by Pilsbry, though in some of my 

 mounted preparations the striae are almost impossible to distin- 

 guish (pi. l,fig. 7; pi. 3, fig. 3). 



The ctenidia number 28-32 on each side. 



Material examined. 



Subgenus Lepidozona Pilsbry, 1892. 



Primarily because of certain differences observed in the radidae, 

 Thiele removes the Lepidozona group of chitons from I schno chiton, 

 and this accomplished finds no recourse except to unite them bodily 



