340 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
largely because of the multiplication by various observers of species 
and genera and even families on small details of difference and 
often on insufficient ground. Of late there seems to be a reaction, 
however, and with it, it is to be hoped that the taxonomy of the 
Irydroids will gradually be simplified. The trouble is probably 
due to the absence of an adequate differential. 
Iliucks (’68) divided the suborder Thecaphora into nine families, 
some of which have been retained, but many have been suppressed. 
These original families were, 1, Campanulariidae, 2, Canqjanuli- 
nidae, 3, Leptoscyphidae, 4, Lafoeidae, 5, Trichydridae, 6, Coppi- 
nidae, 7, Haleciidae, 8, Sertulariidae, and 9, Plumulariidae. 
From this collection of more or less artificial groups it was the 
work of succeeding observers to weed out what was unnecessary and 
to bring together those families which showed natural affinities. 
For example, the first four families were distinguished by very 
imperfect differentials or sets of characters which in the same family 
showed as many variations as in the different families. The divid¬ 
ing line was purely arbitrary, and in the critical examination which 
always follows such arbitrary divisions the weakness was pointed 
out. The first step in the reduction in number of families was 
taken by Levinsen (’93) who brought together in the family 
Campanulariidae all the forms possessing hydrothecae without 
opercula, with the form of a beaker, a bell, or a tube, and with a 
circular mouth opening. This family included most of Hincks’s 
Campanulariidae, and the inoperculate forms of the Lafoeidae. 
All forms resembling the Campanularians in structure of the 
hydrotheca but provided with an operculum, and all the operculate 
Lafoeidae of Hincks were included by Levinsen in Hincks’s second 
family — the Campanulinidae. The chief differential between the 
families, it will be noticed, lay in the presence or absence of an 
operculum. A further reduction in these families was made later 
by Schneider (’97) who showed that the opercular apparatus is 
only a modification of the toothed structure with which, in varying 
degree, nearly all Campanulariae are provided. In other respects 
the Campanulinidae as modified by Levinsen agree well with the 
first family, and Schneider accordingly reduces the two families to 
the one Campanulariidae. He distinguishes, however, between the 
Campanularian-type and the Lafoea-type, forming the subfamilies 
Campanulariinae and Lafoeinae. 
Four of the nine families of Hincks have thus been reduced to 
