135 



protecting their honey from dihition by rain.* While, as he showed, 

 the scale serves to a certain extent as a nectar-guard, its chief use 

 appears to be that of guiding insects which visit the flowers so as to 

 force them to enter all in precisely the same manner. This thought 

 was first suggested by the longitudinal furrow on the lower side of 

 the scale, the two sides projecting downward as far as their width 

 permits. In this form it renders it far easier for insects to enter 

 bodily or insert their heads immediately over the middle line 

 than at either side. The position of the grooved scale, opposed to the 

 sexual organs to which it serves as a guide, corresponds precisely to 

 that of the guiding groove found in Labiatae^ orchids, etc., and 

 the very slight development of this* body in some species is no 

 stronger argument against this view of its function than the 

 slight development of the guiding groove of some plants in the 

 orders just mentioned is against its usual function when well 

 developed. An interesting fact, first observed last spring, is that 

 soop after fertilization has been effected and the stamens have shed 

 their pollen, a slight jar causes the corolla to break loose from the 

 receptacle ; immediately after which the calyx begins to close, forcing 

 the corolla off much as in Verbascum andFer^nua. Several times, I 

 saw this brouglit about by the~visit of wasps belonging to the genus 

 Eumenes, Finding itself, and the corolla to which it clung, suddenly 

 removed from the nectar it was after, appeared to be matter of 

 much s.urprise to the wasp, which visited no more flowers on that 

 plant. This phenomenon doubtless aids in securing the crossing of 

 the different p^lants, in the genera in which it occurs. So far as made 

 out, the process is quite similar in the three genera named ; though 

 whether it is to be attributed to'irritability or not is a matter about 

 which opinions differ. In Rhinathus major and R. minor the corolla 

 is also deciduous; here it does not fall from the receptacle, but tears 

 across near the base, as stated by Rossmann {Bot. Zeitung, i860, 

 P- 217). ^ 



The .inconspicuous lurid flowers, with but a faint variation of 

 color pointing to the nectar, and their unpleasant odor, are less 

 attractive to bees^than to wasps, and it is upon these insects that 

 the flowers chiefly depend for their cross-fertilization. Sprengel 

 states that he has seen a large and a small wasp collect nectar from 

 S, nodosa {Lc, p. 324). H. Miiller found the most frequent visitors 

 for nectar to be Vespa vulgaris^ V. rufa, V. Germanica, V. media^ 

 and V. Jiolsatica. In smaller numbers he found the following bees : 

 Bombus agrorum (once), HaUctus sexnoiaius^ H. zonulus^H, flavipes^ 

 (i, p. 283). The same observer records Bombus senilis as also visit- 

 ing the flowers for nectar (2, p. 267). Darwin speaks of the flowers, 

 of S, aquatica as being fertilized by wasps (i, p. 176; 2, p. 147) 



r 



* The notion formerly was that nectar, being less attractive to insects after 

 dilution, was protected from the rain by these devices. Since, however, the 

 osmotic nature of the secreting process seems probable, it is likely that the neces- 

 sity for maintaining a dense syrup on the surface of the gland is the principal 

 reason for the existence of these nectar-guards, though the reason first mentioned 

 is undoubtedly of considerable importance. Some remarks on this subject may be 

 found in the Botanical Gazette for Nov., 1881, p. 287, , 



^^4^^^_1,^^^ U^^^^^t.^^^ 



'6^.A Jt # 



^i ;v. 



v^ 



