306 
witschia and the entire coat of Physostoma, Trigonocarpus, 
Mitrospermum, Bennettites, of the Cycadean seeds, of 
Ginkgo, and of the seeds of the Pinaceae and Cupres- 
saceae are homologous organs, always showing traces of 
units. These organs only (about the integuments of 
Ephedra and Gnetum 1 will speak presentiy) I should 
like to entitle with the name integument. 
We may not yet fully determine the value of the cupule 
of Lagenostoma and Sphaerostoma, but it is most probable 
that they represent entire sporophylls, at least fused pinnae 
of it, like the disk of Bennettites and the arillus of Ginkgo. 
The cupule of Taxus and Torreya, as well as the 
perianth of Gnetum and Welwitschia consist of two fused 
sporophylls as is the case in Ephedra. Finally the epi- 
matium of Podocarpus is an entirely new formation without 
any connecting link in other fructifications. Perhaps we 
find a homologous case in the outgrowth of the sporo- 
phyll at the base of the ovule in some Cycads. But as 
long as no better explanation can be given, the name 
epimatium ,,Excreszenz des Carpides” (117 p. 16), is the 
best term we can use. 
We have seen that in the ovules of the Taxads the 
endotesta is loosened from the stony coat in its upper 
part, and there forms a distinct layer on the nucellus. 
This may be a transitional stage to the case represented 
by ÆEphedra and Gnetum, where two integuments exsist. 
The outer integument consists of a sarcotesta and a stony 
layer, whereas the inner one is very thin and may be 
compared with the endotesta in other seeds. 
It always has been supposed that amongst the Gnetales 
at once a second integument has appeared which has 
been persistent in the Angiosperms. But now there has 
become a closer relation between the ovules with one- 
and those with two integuments, on account of this 
connecting link, the ovules of the Taxads. However, we 
