MR. G. MAW ON THE LIFE-HISTORY OF A CROCUS. 859 
treating of the geographical distribution of the genus and of the 
presence of special characters related to geographical association. 
The Filament, which springs from the base of the three outer 
segments, is generally white or yellow ; it often partakes of the 
colour of the throat; and in several deep-purple species the fila- 
ment is purplish. In afew species, notably in those with white 
anthers, the filament is essentially orange; and in C. cyprius it is 
scarlet. The filament is generally a little papillose; and in two 
or three species, e. g. O. pulchellus and C. Tournefortii, is densely 
covered with hairs. 
The filament is generally about half the length of the anther ; 
in C. minimus it is as long, and in C. Boissieri much longer than, 
the anther. The relative length of the filament and anther is 
constant within each species, and is of use for specifie diagnosis. 
The Anther.—There are few special specific characters in the 
anthers that need be referred to, except that in seven or eight 
Species the anthers are white, in one species (C. Crewei) dark 
chocolate, and in all the remaining species orange. The colour 
of the anthers, unlike the variable colour of the stigmata, is con- 
stant within each species, and, as far as my observations go, never 
Varies excepting in the occasional occurrence of dark-chocolate 
tissue near the base. The only other special specific character I 
have met with is in the suddenly divergent anther of C. aureus, 
which distinguishes that species from any other. 
Pollen.—The pollen-grains have also their special specific 
characters, varying much in size, from £1, to 51; of an inch in 
diameter in different species. Within each species the size is 
constant; they are regular spheres, excepting only in C. sativus 
and the allied species, in which they are variable in size and 
also of irregular form. The grains of the great majority of spe- 
cies are orange, those of the white-anthered species pale cream- 
colour. In about half the species they are covered with minute 
Papillge, and in the remainder are glabrous. In the pollen-grains 
of C. aureus and its allies Mohl, followed by Etlgeworth, described 
the presence of a spiral coil or double loop, which, Mr. Carruthers 
has ascertained, is due to the thinning or channelling of the outer 
coat of the grain. I have also detected it in C. carpetanus, C. 
ancyrensis, and one or two other species. 
Stigma.—The stigmata present a great variety of types, so 
diverse in their character that I am unable to adopt the three- 
fold classification, based on the stigmata, proposed by Mr. Baker. 
lt is beyond the scope of this paper to particularize the various 
