PART III. | THE MECHANISMS OF FLOWERS. 413 
long are able to reach the honey. Bees touch the longer reproduc- 
tive organs with their heads or with the base of their proboscides, 
and the shorter with the maxillz, which serve to sheath the pro- 
boscis ; thus they regularly perform ‘legitimate cross-fertilisation,’ 
The flowers are very conspicuous and very rich in honey, and 
appear at a season when they have few rivals; they are therefore 
very much visited by insects, and have become sterile when 
fertilised with their own pollen or with pollen from another flower 
of the same form. Hildebrand (342) found by experiment that 
when he fertilised a flower of either form with its own pollen, or 
with pollen from another similar flower, it was completely sterile ; 
but when fertilised with pollen from a flower of the other form, it 
was as productive as in the wild state. He found on investigating 
wild plants in fruit, that in some cases the oldest flowers on the 
plant, and almost constantly the terminal flowers of each shoot, bore 
no seed; the former fact he explains by the absence of the proper 
insects early in the season, the latter by imperfect nourishment of 
the terminal parts of the shoots. 
Darwin’s experiments (No. 164, p. 103) on this plant led to a 
result: different from Hildebrand’s. He found that illegitimately 
fertilised long-styled plants were highly fertile, producing three 
times as much seed as Hildebrand’s wild specimens bore; and that 
even when self-fertilised, a few seeds were produced. Hildebrand 
endeavoured to explain this great discordance by the fact that the 
plants which he experimented on were kept in pots in the house, 
while Darwin’s were grown out of doors. 
Visitors: A. Hymenoptera—Apide: (1) Anthophora pilipes, F. ¢ 9 
(19—21), very ab., s. andc.p., sucking now on Primula elatior, now on Cory- 
dalis, now on Pulmonaria, without restricting itself long to the same species ; 
- (2) Bombus hortorum, L. ? (21), very ab., s., and keeping to the same species 
of flower ; (3) B. lapidarius, L. 9 (12—14), s. ; (4) B. senilis, Sm. 9 (14—15),s. ; 
(5) B. agrorum, F. 9 (12—15), ab., s. ; (6) B. silvarum, L. 9 (12—14), s.; (7) 
B. Rajellus, Ill. 9 (12—13), s. ; (8) B. terrestris, L. 9 (7—9), s.; (9) B. pra- 
torum, L, 2 (11—12), s., distinctly prefers Pulmonaria, leaving the flowers of 
Primula elatior untouched ; (10) Osmia fusca, Christ. (bicolor, Schrank) ? ¢ 
(8), c.p. and s.,ab. This species feeds itself and its young almost entirely on 
the honey and pollen of Pulmonaria. I have never found its nests (which 
are made in snail-shells, those of Helix nemoralis at Lippstadt) except where 
Pulmonaria was growing plentifully. (11) O. pilicornis,Sm. ¢ 9,s.andep. I 
was the first to find this bee on the continent of Europe. I have found it ex- 
clusively on flowers of Pulmonaria, at Rixbeck near Lippstadt ; it occurs singly 
among numerous examples of the foregoing species, with which it agrees in the 
manner of tending its young. (12) O. rufa, L. ¢ (7—8),s. In most of these 
bees I have directly observed pollen upon the maxilla. B. Diptera—(a) 
