282 
Transactions of the Society. 
in which way the numeration is made is not a matter of much import- 
ance, though as there are some cases in which it is better to give the 
position of the plates in the entire wall, I have done so, adhering to 
the plan adopted in my previous papers, and therefore, to compare 
Levinsen’s figures with mine, his must he doubled. 
On the lateral wall there may be two plates, as in id armata , 
papyrea, nobilis, pisciformis , serrulata , ovoidea, militaris , Euthyris 
bombycina . 
Or four plates, as in F. biseriata, crassa , dissimilis , elegans. 
Or six , as in F. membranaceo-truncata, Barleei, oblonga , separata , 
cyathiformis, foliacea, carbasea, securifrons, Euthyris episcopalis. 
Or eight, as in Euthyris obtecta. 
Or ten, as in F. tenella, papyracea, reticulum , denticulata, denti- 
gera ; or over ten, as in F. spimdigera and cribriformis. 
We need not be surprised to find that there is not absolute regu- 
larity in the number of these plates, and naturally the regularity is 
greatest in those species with the fewest plates ; so that, while in those 
with two we may speak of the number as constant, this is by no 
means the case with those having a larger number. 
There are two ways in which irregularity may occur. Taking 
as an example Membranipora membranacea, there are usually four 
lateral plates, but sometimes one of these plates is seen to be replaced 
by two smaller ones, thus three large and two small, or some such 
irregularity ; but even then it is generally easy to say what is the 
typical number, but should this not be the case, the examination of 
a few zocecia will show what must be considered as typical. The 
other form of irregularity occurs more with the many plated species, 
for then sometimes the plates may be said to occur at equal distances ; 
and as irregularity in the shape of the zooecia means that the length 
of the wall may vary considerably, the number will correspond with 
the length. This was pointed out by Yigelius * for F. membranaceo- 
truncata, and he says the number for each wall is fairly inconstant. 
However, he figures | what I should consider a typical zocecium show- 
ing six lateral and three distal plates. Levinsen gives the range 
within which variation takes place. 
The results of the comparison of 29 species are given as a table 
to facilitate reference. The genus Euthyris, comprising E. obtecta , 
episcopalis , and bombycina has been included. E. episcopalis has 
been called Onchoporella, but as to a new genus being required, I am 
very doubtful. F. biseriata v ill probably have to be removed from 
Flustra. However, the object of this paper is not to define the limits 
of Flustra, but to show that there are additional characters which 
may be used in studying their relationships. 
Although there is this interzooecial communication through the 
* Yigelius, W. J., Morph. Unters. fiber Flustra membranaceo-truncata , Biol. 
Cent. Blatt., iii. p. 708; and ‘Die Bryozoen des Willem Barents,’ p. 21. 
f Yigelius, Bry. des W. Barents, pi. i. fig. 9. 
