STUDY OF SPONTANEOUS HYBRIDS IN EUROPEAN FLORA. 153 
considerably from those of the lateral stalks or of the lower branches. 
Fertility relative. In this genus it is easy to observe the mongrels of 
which I have already spoken above. In certain peat-bogs of the Jura, 
I have found it impossible to separate the abundant hybrids which I have 
collected, the forms of reversion being numerous, thus making it very 
difficult to distinguish the different forms pointed out by authors. 
Carduus.—Crosses numerous; individuals abundant. Relative sterility. 
Centaurea.—Crosses numerous ; number of individuals very variable ; 
sterility frequent. 
Hieracium.—Hybrids and mongrels very frequent. Fertility relative. 
Individuals springing from crossings bear evidence of the forms from 
which they originate ; the same is true of the mongrels (métis). 
GENTIANACE®.—Gentians are easily crossed, but the capsules bear 
few seeds, and the hybrids produced are not numerous. 
PrimuLAcez#.—The genus Primula is one of the most propitious for the 
study of hybridisation. Certain crosses are yery abundant, x P. digenea 
(vulgaris x elatior), x P. variabilis (officinalis x vulgaris); others are 
much more rare, as x P. media (elatior x officinalis). Where P. vulgaris, 
elatior, and officinalis are crossed together there always results a great 
quantity of hybrids of the first two forms. P. media is on the contrary 
an exception, which may be explained by the fact that insects which visit 
P. elatior more rarely despoil P. officinalis. The small throat of these 
two species is perhaps also an obstacle to their visits. In Primula the 
capsules frequently prove abortive or enclose but few well-developed seeds. 
The calyx is the organ which provides the most definite characteristics 
by which to recognise traces of descent. 
The genus Soldanella and Androsace, in proportion to the small 
number of species of which they are composed, provide a large number 
of hybrids. 
SCROPHULARINE®. Verbascum.—With Verbaseum as with Primula 
hybridisations are easy; the results produced are numerous and remark- 
able for their height, which prevents their escaping observation. Sterility 
is almost universal in this genus. 
Huphrasia and Pedicularis.—Cross-bred forms are numerous, with 
results varying in number, some very numerous; others represented by 
almost isolated individuals. 
' Lasrara.—The genus Mentha, on which my eminent colleague 
M. Malinvand, President of the Botanical Society of France, has con- 
tributed an article, is one of the most interesting from our point of view. 
Researches, especially those lately undertaken, have shown that hybridisa- 
tion may often be observed in the following genera ; Lavandula, Salvia, 
Thymus, Marrubium, Stachys, Galeopsis, Prunella, Teucrium. 
CHENOPODIACEs.— Hybridisation rare. 
Potyconacez. Rumex.—Hybridisation frequent, results numerous ; 
fruit more or less abortive. 
Polygonum.—Crosses frequent, results numerous, sterility less marked. 
CUPULIFERE. Guercus.—Hybridisation frequent, results keeping the 
characteristics of the form from which they have originated. 
SauicinEx. Saliz.—Hybridisation very frequent, even between species 
differing much. The frequency of hybridisation is much favoured by the 
separation of the sexes and the presence of nectaries, and the often intense 
