70 
BULLETIN 106, UNITED STATES NATIONAL MUSEUM. 
The bryozoa are reduced for food to planktonic organisms; it is probable that 
they thrive particularly in places where a brisk current exists. The straits and 
narrow passages are ordinarily rich in bryozoa. (Nordgaard, 1912.) They feed 
upon diatoms and radiolaria (fig. 15, F). 
The moving sands with large siliceous elements are not favorable for the 
habitat of bryozoa and there one can scarcely encounter a single example. The 
rocky facies, on the contrary, lend themselves very well to the establishment of 
colonies of bryozoa which swarm there. The muddy bottoms are very poor in 
bryozoa (Guerin-Ganivet, 1912). 
The bryozoa are rather numerous in the shell sands. They are numerous in 
the zone of the mud sands with microcosms (25-60 meters) : the shells serve as 
substratum to the numerous incrusting bryozoa, whereas the erect bryozoa more 
generally spread over the tunic of ascidians (Calvet, 1902). 
The zoarial articulation is in rapport with the mobility of the habitat and with 
the zooecial fragility which requires some protection. All articulated species are 
then commensals of the great marine meadows. 
PRINCIPLES OF CLASSIFICATION OF THE CHEILOSTOMATA. 
The principles of classification of this order are still imperfect in spite of the 
quite extended research of several students. Formerly the classification was based 
on purely zoarial features, but in the latter half of the nineteenth century the 
zooecial characters Avere more closely studied, especially by D’Orbigny, Smitt, and 
Hincks. The latter author considered especially the form of the aperture, in 
other words, only the hydrostatic system. In 1888, and again in 1903, Jullien 
established the systematic set of characters for consideration. These are as follows 
in diminishing order of importance : 
Essential characters; general morphology (order) ; form of the frontal wall 
(sub-order) ; form of the aperture and of the operculum (family) ; presence of 
cardelles, occurrence of lyrula, and finally ovicells and radicels. 
Secondary characters or specific; frontal punctations, avicularia, and vibracula. 
In 1900 Canu wrote that every family ought to be based on an anatomical 
peculiarity common to all its members and fixed in an uninterrupted series of 
descendence. He established genera according to the variations of this anatomical 
peculiarity and according to the divergence of its evolutionary characters. This 
was the perfection of the ideas of Jullien, but the partial application made by 
Waters on the opercula and the avicularian mandibles did not appear always to lead 
to universal results nor to the establishment of very natural genera. 
We noAV believe that other principles are better. In the bryozoa, as in other 
living beings, the form is only the result of functions; therefore in the study of 
the morphological variations of the organs we now substitute that of their physio- 
logic functions. Our studies are therefore always directed toward the discovery of 
functions which modify the skeletal form. 
