The Coccidae of South Africa. 
85 
the greatest value. For instance, the majority of species have well-defined 
marginal spine areas, while in a few cases such are absent (Figs. 22-33). 
In some again the gland-pores over the whole surface of the body are 
uniform in size, in others the ventral pores are small and simple, while 
those of the dorsum are large and disc-like. Further, some species have 
subcutaneous tube-glands restricted to the posterior segments, in others 
they may be uniformly distributed or entirely absent. In Bhizoecus there 
are peculiar grouped glands, in groups of three, crowded together and pre- 
senting a propeller-like appearance. The character of the spines and 
bristles on the derm varies, the conical spines of the marginal areas being 
stout, or slender, or entirely bristle-like, or absent. 
Unfortunately I have not seen stained material of a sufficient number 
of species to be able to classify them according to the different types, but 
I refer to the subject in the hope that some one more favourably situated 
than I am may undertake to do so. 
{k) Setae of Anal Bing and Anal Lobes. — The only mention of this 
character in the majority of descriptions is the number of hairs which the 
anal ring bears. All the series which I now include in the Pseudococcini 
have the anal ring with six hairs. It was observed that the length of the 
anal ring hairs remained fairly constant throughout all the females of the 
same species, but this did not furnish a satisfactory specific character in 
itself, because there was not sufficient difference between the lengths in 
different species. The anal lobes are furnished with conical spines and 
glandular pores, and each bears one long seta. These setae remain fairly 
constant in length for the species, so that the comparative length of the 
setae of the anal lobes with those of the anal ring forms, I think, quite 
a good character when taken in conjunction with the antennal segments 
and the legs. 
Intracellular Synibionts. — Very little is known concerning the intra- 
cellular symbionts of the Coccidae ; but, apart from the intense interest 
attached to them — owing to their relation with metabolism — they may 
throw a very important light on the subject of relationship. This may 
not indicate specific differences in many instances, but regarding the 
larger groups and their derivation I feel that a great deal may be 
anticipated. 
The following seven species have been described from Coccids : — 
1. Lecaniascus polymorphus Moniez, 1887, from Lecanium hesperiduni. 
2. Saccharomyces apiculatus var. ijarasiticus Lindner, 1895, from 
Saissetia hemispherica. 
3. Oospora saccardiana Berlese, 1906, from Geroplastes rusci. 
4. Saccharomyces pseudococci farinosi Sulc, 1910, from Pseudococcus 
farinosi D.G. 
5. Coccidomyces Buchner, 1911, from Lecanium conn. 
