LORICULA. 88 



there are about twenty-one scales, their numbers obviously depending on the age and size 

 of the individual. There is one more scale under the second latus (fig. 4, restored figure), 

 than under the first latus, and one more under the first latus than under the scutum ; hence 

 the summit of the peduncle is obliquely truncated, being lowest at the rostral end : in this 

 respect there is some resemblance to the genus Lithotrya. The scales in the three lateral 

 rows resemble each other pretty closely in outline : they are transversely elongated, and 

 arc together about as wide as the three valves of the capitulum ; they are much longer 

 than the scales of either end row. The first row of these elongated scales lies directly 

 under the scutum, and the other two under the two latera ; so that the lines of junction of 

 the three rows of peduncular scales, and of the three valves of the capitulum, correspond. 

 The scales in the middle row are rather longer than those on either side of it, are pointed 

 at both ends, and have their upper margin very flatly arched and almost square-edged. 

 The scales in the row under the second latus are rather wider than it, projecting (which is 

 important) beyond its outer edge ; their upper margins are square-edged ; their outer ends 

 blunt and truncated; their inner ends pointed. The scales of the third row under the 

 scutum are rather less wide than it, and do not reach so far as its outer or rostral angle ; 

 their upper margins are arched ; their outer ends bluntly rounded, and their inner ends 

 pointed. 



The two end (that is, the rostral and carinal) rows of narrow scales remain to be 

 described : those at the rostral end (fig. 3) are as high as the larger lateral scales, but only 

 about one fourth as wide : in shape they are almost a rectangular oblong, with their 

 upper ends a little rounded, and the outer (with respect to the longitudinal axis of the 

 animal) basal angle a little produced ; hence the two lateral margins of the scales of 

 this rostral row do not quite match each other ; consequently, to make the animal symme- 

 trical, there must have been a corresponding approximate row of small scales on the other 

 side of the medial line. The straight inner (both sides of the peduncle being supposed 

 to be present) edges of the scales of the row just described, extend rather beyond the 

 occludent margin of the scuta. The scales in the carinal row, at the opposite end of the 

 peduncle, are not above half the width of those of the rostral row : they are of nearly 

 the same shape, but their upper ends are more pointed, and their outer (with respect to 

 the medial longitudinal axis of the animal) basal angles more produced ; their straight 

 inner margins projected considerably beyond the carinal edge of the second latera : it is 

 more obvious in this case than in the rostral row, that there must have been a second 

 adjoining row of small scales on the other side of the carinal medial line. The small 

 scales of the carinal and rostral rows differ from the others in their inner (that is, close to 

 the medial axis), basal angles, not crossing each other; so that the peduncle could have 

 been divided in a medial rostro-carinal plane, without cutting through a single scale. 

 Their outer basal angles, on the other hand, intersect the ends of the adjoining large 

 lateral scales, like these latter intersect each other, the lines of intersection between them 

 being straight and corresponding with the junctions (as already stated) of the scuta and 



