450 THE FERTILISATION OF FLOWERS. [PART III. 
Delpino (178) seems to have examined the large-flowered form 
only, and Axell (17) the small-flowered ; for the former describes 
E. officinalis as proterogynous, and the latter as proterandrous. 
Visitors: A. Hymenoptera—Apide: (1) Bombus agrorum, F. $; (2) B. 
pratorum, L. %; (3) Apis mellifica, L. $ ; (4) Nomada lateralis, Pz. 9, all s. 
B. Diptera—(a) Bombylide : (5) Systcechus sulfureus, Mik. (Sld.) ; (6) Syr- 
phide: (6) Syrphus sp.; (7) Melithreptus teniatus, Mgn., all s. Additional 
lists of visitors to the large-flowered form are given in No. 590, IL, and 
No. 609. 
Huphrasia salisburgensis, Funck.—The flowers resemble in size 
the small-flowered form of £. officinalis, and their arrangements 
for cross-fertilisation resemble the large-flowered form of that 
species. The corolla continues to grow, carrying forward the 
stamens after the style has stopped growing, and in absence of 
insects self-fertilisation takes place in the end. The visitors are 
bees and Lepidoptera (609). 
333. PEDICULARIS SILVATICA, L.—In this plant, as in Rhinan- 
thus and Melampyrum, the opposite pairs of anthers lie with their 
edges close together,—and all four unite to form a quadrangular 
pollen-receptacle (3, Fig. 155) which is inclosed in the hooded upper 
lip. The mechanism by which pollen is shed upon the insect- 
visitor is here more complex and more rich in adaptations than 
ever, and more difficult to understand in its details without direct 
observation of its action. Sprengel (702) and Hildebrand (346), 
neither of whom observed insects in the act of fertilising the 
flower, have very naturally overlooked many of its most interesting 
adaptations. Dr. Ogle (632), who watched the flowers in the open 
air and saw them visited by humble-bees, gives an accurate 
account of the process of cross-fertilisation ; but many adaptations 
have escaped even his notice, and some he has interpreted 
wrongly. It is therefore still worth while to describe the whole 
structure of the flower minutely. 
Honey is secreted by the green, fleshy base of the ovary, and 
lies at the bottom of the tube, which is 10 to 14 mm. long, laterally 
compressed, and so narrow that a humble-bee can only insert the 
forepart of its head. At the height of 10 to 14 mm. the tube 
divides into an upper lip, which both in form and in direction is 
simply a continuation of the tube, slightly wider and open in 
front, and an under lip, whose base (8 to 5 mm. long) is applied 
to thie upper lip so as to close up the lower 3 to 5 mm, of the 
