222 MR. I. H. BURKILL ON SOME VARIATIONS IN 
(C) From a warm sunny cinder-bed in the Cambridge Botanic 
Gardens (shaded after June 20th). The plants were examined 
almost every week throughout the whole time in which they 
flowered. 
(D) From the Pinetum in the Cambridge Botanie Gardens. 
The plants grew in the dense shade of trees of Cedrus, Thuja, 
and Tazus, and received no direct sunlight. They were exa- 
mined weekly almost to the end of their flowering period. 
(E) and (F) From two pots grown from seed in the ‘ Inter- 
mediate (7. e. subtropical) pit’ in the Cambridge Botanic Gardens. 
In E the soil was rich; in F it was mixed with an equal quantity 
of sand. The plants were examined three or four times a week 
during the time in which they flowered. Being under glass, they 
obtained slightly less sunlight than those on the cinder-bed, but 
received more heat. The pit is kept at an average temperature 
of 60°-65° during the day, and 55° at night. 
(G) From gravelly ground, slightly shaded. Only two exa- 
minations were made, both in March. 
(H) From damp but sunny ground on Coe Fen, Cambridge. 
One examination only, which was made in March. 
(1) From the cinder-bed above mentioned. One examination, 
made in August, of the first flowers of a young growth. 
(J) and (K) From two pots grown in the open. The same 
soil used as in E and F; K contained 50 per cent. sand, as in F. 
All the seed sown in these four pots (E, F, J, and K) was gathered 
at the same time and from the same place. The roots of these 
plants were able to spread more freely than those of E and F. 
Unfortunately an accident caused the death of the plants of 
both J and K, and prevented the examination of the later flowers. 
(L) From the highest point of Seamer Moor, Scarborough 
(600 feet). The locality is sunny. A single examination made 
in April. 
Extent of Variation —Hermann Müller * has noted the varia- 
tion in the number of stamens, remarking that the most weakly 
plants have the fewest stamens: he points out the frequency 
with which the stamens are reduced to three. A glance at 
Table I. shows at once the predominance of this number, no less 
than 41:70 per eent. of the flowers having three stamens; while 
* “ Weitere Beobachtungen ü. Befruchtung der Blumen durch Insekten.” 
Verhandl. d. naturhist. Vereins d. preuss. Rheinland u. Westfalens, xxxvi. 
p. 228 (1879). 
