Echinoid Tests, Terminology. 
63 
proposer of the term, A. Gras («Ce scrobicule est parfois borde ä sa circonference d’un 
cercle plus ou moins saillant et serre de tubercules, cercle que nous appellerons scrobiculaire ». 
1848, p. 301). When a clear intertubercular tract intervenes between the scrobicular rings 
of a meridian, then the scrobicules are termed distinct and separate. When the 
scrobicular rings are complete, each in itself, but separated by no such space, the scrobicules 
are distinct but a d j a c e n t. When the scrobicular circles remain complete, but when 
the rings meet so that there is only one line of tubercles between the scrobicules, then the 
scrobicules are contiguous. When the scrobicules are so close that both their rings and 
circles are incomplete and one scrobicule merges in another, then the scrobicules are 
confluent. 
The physiological significance of the undercut neck, the platform, the parapet, and 
the crenelation is not at once apparent. The internal sheath connecting the boss with the 
radiole was formerly supposed to consist of elastic ligament-fibres, but J. von Uexküll 
(1899, Zeitschr. f. Biol. XXXIX, p. 73) has shown that it is muscular and serves to hold 
the radiole rigidly erect; it passes from the basal terrace, over the crenellae of the boss, 
and is attached to the base of the radiole below the occasionally crenelate collar of the 
latter. Therefore neither the crenellae of the boss nor the grooves between them serve as 
points of attachment for the internal muscle-fibres. They may, however, hold the fibres in 
a straight line, and thus %check torsion of the radiole. The space beneath the internal muscle- 
sheath, formed by the undercutting of the neck and excavation of the platform may be due 
partly to economy of material, while it may also facilitate the movement of the acetabulum 
over the mamelon, may allow room for the swelling of the muscles when the radiole is drawn 
over, and finally may provide a soft cushion to save from injury by accidental pressure or 
blows the nerve-ring which lies at exactly this level in the epidermis covering the union of 
tubercle and radiole. 
The remaining Ornament of the interambulacra consists of tubercles or granules, of 
which the former are distinguished by being mamelonate. These tubercles are s e c o n d' 
ary, tertiary, and so on according to their size. The principal scrobicular tubercles 
are almost invariably larger than any others, and are therefore secondary. Secondary tubercles 
are frequently scrobiculate; they may also be perforate and crenelate. Granules 
are devoid of mamelon, Perforation, crenellae, and scrobicule. The term «miliaries» («miliary 
tubercles» or «miliary granules») is often used, but there seems considerable uncertainty in 
its application. Duncan (1889, p. 298) says miliaries are «very small tubercles incomplete 
in their division into parts», while «granules are more or less nodular projections of the 
test» : this is far from clear. Wright (1857, p. 14) distinguishes granules from tubercles 
in the same way as is done here, and then opposes them to miliaries in the following 
manner: «granules are . . . scattered more or less regularly, and distributed over different 
parts of the plates of the test», while «miliary granulation is formed by a number of 
small granules closely set together» in the perradial tracts or the interradial tracts, and 
these tracts he calls the miliary zones. Since, however, ordinary granules may, as stated 
by Wright, occur in the perradial tract of Cidaris, and since a close-set irregulär granu¬ 
lation is often found in the adradial tracts of the interambulacra, Situation fails as a criterion. 
A. Agassiz (1874, p. 636) distinguishes «miliary tubercles» from «granules», but defines 
neither; it is, however, granules and not miliary tubercles which, according to him, occur 
in the miliary zones. To escape the confusion exemplified by these quotations, it is time 
that we returned to the simple conception of C. DesmoulinS (1835, p. 15). For him all 
eminences bearing skeletal processes (radioles, pedicellariae, &c.) were «tubercules», and of 
