VARIATION IN ORCHIS MACULATA 
91 
no longer found : the dark Howers grow at the north end and the 
light at the east end, the two groups merging into each other, 
but no very light flowers are found at the north end and no very 
dark flowers at the east end. 
On June 15 , igoo, about a week after the orchid first 
appeared, 1 gathered forty specimens at this spot, and found that 
on the basis of flower colouration they could be divided into eight 
groups, of which the first was nearly white and the last a darkish 
purple, while the other groups were of intermediary shades. 
The two extremes were so unlike that, even to a botanist, at the 
first glance, they seemed to belong to two species. About a 
fortnight later, when the orchid had reached its full development, 
I gathered thirty-five specimens. These showed no characters 
not noted on the first occasion, hut the detailed remarks which 
follow refer only to the results of the second collection. 
On the basis of flower colouration I divided the specimens into 
four groups. Group A. (ten specimens) showed flowers that 
were almost white, the petals, apart from their small violet spots, 
only very faintly, or not at all, tinged with violet ; Group B. 
(eleven specimens) showed petals of a distinct, though very 
pale violet, while the spots were darker; Group C. (eleven 
specimens) showed petals of a pale purplish violet, while the 
spots were still darker and tended to run into wavy bands of 
colour; Group D. (three specimens) showed still darker purplish 
petals, sometimes almost as dark as the bands. We thus have 
two groups that may be called light, and two that may be called 
dark. 
The pigmentation of the stalk (invariably short at the spot 
investigated) ran without exception pari passu with that of the 
flowers : while almost absent in A., it became a dark brownish 
purple in the upper part of stalk in D., as also the bracts which 
in A. were very pale green. 
The leaves of Orchis macnlata are spotted with blackish pig- 
ment, whence the plant’s name. This pigmentation is verj' 
variable, and at first 1 was inclined to think it had no relation to 
depth of flower colouration. 1 found, however, that though B. 
showed some of the most deeply pigmented leaves, all the pale- 
leafed specimens were in A. and B., while C. and D. showed no 
very pale-leafed specimen. 
With these marked colour variations were associated certain 
morphological variations. The lip of this flower tends to form 
three lobes, but while in some of the specimens the middle lobe 
was on the same level as the outer lobe, in others the middle 
lobe formed a tongue projecting considerably beyond the other 
two lobes. In the light groups, A. and B., the lip was of 
the first type, with occasional very slight tendency to the second 
type in B.; in the dark groups, C. and D., it was always of the 
second type. 
Another character of this orchid is its honey-like odour. In 
A. I found this odour distinct in every specimen, as well as in 
