Sep., 1890. 
weismann’s theory of germ-plasm. 
207 
spores, but the species are now extinct. Nevertheless we 
have two small families of fern-like plants mutually very 
distinct, which, in spite of their great difference, are still 
grouped together under the absurd name of Rliizocarpese. 
These are the Salvineae and the Marsilese, in which are formed 
two kinds of spores entirely different in nature; and the 
same peculiarity is again met with in the case of the great 
subdivision of the Vascular Cryptogams, namely the Lyco- 
podiaceas ( Dichotomece ). Here also are two very different 
families, the Selaginellae and the Isoeteae, in which two kinds 
of spores are produced. I cannot here repress the remark 
that it harmonises little with Darwin’s views when we see 
repeated in three different classes of the vegetable kingdom, 
with otherwise similar spores, a phenomenon so important as 
the production of two kinds of spores, with their consequences. 
Certainly it cannot be explained by Natural Selection in the 
struggle for existence.” 
To those who are familiar with the complicated structure 
of the macrospores, and their relation to the microspores, 
this will appeal strongly. It is sufficient for me to say that 
they are very highly developed structures, almost as much so 
as the ovules and pollen grains of the Conifers, and stand in 
the same relation to each other as they do. In this case it is 
even more difficult to believe that a parallel series of varia¬ 
tions occurred leading to identically similar structures in 
these widely separated plants. It may be argued that it is no 
more improbable than that carnivores, though but dis¬ 
tantly related, are found with similar habits widely distributed 
over the world, and separated by oceans, but this cannot be 
called a parallel of these two cases, for in tracing the descent 
of the cuckoo or of the Heterosporous Cryptogams, those 
peculiarities that so prominently distinguish them disappear 
before the point is reached where the two lines converge, 
which would not be the case with the Carnivora. 
The Brine Shrimp (Artemia) is partlienogenetic, or only 
very occasionally reproduces itself by sexual means ; the pro¬ 
portion of males to females being 1-3,000, so that if variation 
can only occur during a sexual generation the chances of a 
variation in any one direction are very small; yet the Artemia 
Mulhousenii , which lives in water containing 25 per cent, of 
salt, can be caused, by diluting the water to 4-5 per cent., to 
change to A. salina ; by further diluting to freshness, it 
changes to a form belonging to a different genus, Brancliypus, 
a much larger creature, with quite different structure both 
internal and external. Now if the chances of variation are 
one in three thousand, and the chances of a favourable one 
