INTRODUCTION. CXX111 



below tlu- aperture is tubular or obcouic; the Eschurimi, 

 with calcareous cells, the aperture of which about equals 

 in size tin- opereuluin, no membranous area being left; 

 and tin- ('<//<'jjorinu, with calcareous cells more or less 

 \vrtical to the plane of the colony, with a terminal aper- 

 ture, and irregularly heaped together. The families and 

 genera are constructed in harmony with the principles 

 whieh have been briefly indicated. 



It is impossible to estimate too highly the thoroughness 

 of research on which Prof. Smitt's classification rests, and 

 the important contribution which he has made towards a 

 natural system, however much we may be disposed to 

 dissent from some of his results. 



In his great work on the Embryology of the Polyzoa *, 

 Barrois contends that in classifying these animals account 

 should be taken of all the forms in the life- series toge- 

 ther, and sketches an arrangement suggested by the study 

 of the larvae (p. 250) , for comparison with that of Smitt, 

 from which it differs in some important respects. Our 

 knowledge of the larval forms, however, is as yet too frag- 

 mentary to allow of our building much upon it ; Barrois's 

 sketch is quite hypothetical, and, though interesting in 

 itself, can only be accepted as a suggestion requiring the 

 confirmation of further research. 



In the present work the Polyzoa are ranged under the 

 subclasses instituted by Ray Lankester, and based on the 

 cliaiaeUT.s of the lophophore. Most of them are Holo- 

 Iminchiute, and have the tentacles in a continuous series. 

 \ single genus only exhibits the Pterobraiichiate struc- 

 ture, in which the Molluscau character is more pro- 

 nounced, and the lophophore is broken into two distinct 

 * Hii-ii. Mir IVinbryologie dee Brvoioaires,' 1877. 



