18-1- The Ariierira/i Genloi/isf. Marcli, 1895 
'I'lic (lia^i'nin on I'liitc 11, iigures 1, 2, H, 4, o, represent 
stagX'S in the developinent of Farosilcs forhcsi^ syinnietrically 
considered. 
Two groujis may l)e distinguished among tlie colonies just 
discussed. In one, the initial cell is strongly upriglit, with 
hut a slight surface of attacliment. From tiiis there results a 
corallum elongated in form, and much resembling a Cyatho- 
phylloid in appearance. No specimen with this characteristic 
has been observed still attached. 
The other form occurs in greater numbers. Individuals 
representing it are found attached to crinoid columns where 
they ultimatel}" form a ring about the stem. They have also 
been observed upon ramose forms of Br^^^zoa. The initial 
<a>ll, as well as the five or six corallites next ])roduced, are 
prostrate upon the surface of support, becoming resurgent 
only when the first l)ud is formed. This variation seems to 
be the direct result of conditions which surround the coralla. 
As the acute-conical shapes were presunuibly in a place less 
favorable for growth, they assumed an upright position and 
direction. The other form, the explanate almost incrusting 
one, being thrown by chance upon an advantageous place, as- 
sumed temporarily a resupinate mode, of growth. 
Thirty-eight coralla have been f(uiiid nearl}' entire, and i]i a 
state of good preservation. The development from the initial 
corallite through two or more generations of buds can thus 
be traced out with considerable certainty. Of these coralla, 
three were younger than the completed second stage. Of the 
reinaining thirty-five, four jjroduced three, and thirty-one 
produceil four peripheral buils as the second stage, and in ev- 
ery instance, these buds iippeared on the dorsal si(k^ Those 
features, therefore, which constitute perhaps the only new 
and important points brought out in this paper seem to be 
well established. 
Irregularities in the development just outlined are not so 
great as to preclude reduction of the whole to a general sys- 
tem. By comparing irregular colonies with more symmetri- 
cal ones, an ideal law nuiy be detected toward which all are 
tending. And so the word law may l)e used in connection 
with tlie developnuMit of coralla, but only in this sense, that 
certain buds habitually appear in certain symmetrical posi- 
