60 MR. J. C. WILLIS ON THE 
folia and P. divaricata) have small, pale-coloured flowers, massed 
in dense cymes over which insects can crawl without flying from 
flower to flower. The other three species, all belonging to the 
same subgenus (Whitlavia), have large brightly coloured flowers, 
separated from one another upon the axis of the cyme, 80 that 
insects must fly from one to another. The former species are 
best adapted to, and seem often to obtain, self-fertilization ; the 
latter are better adapted to across. All are dichogamous, but the 
larger-flowered species more so than the rest. In all the honey is 
secreted below the base of the ovary and protected by appendages, 
which in P. tanacetifolia ave attached to the corolla, in P. Cam- 
panularia, Whitlavia, and Parryi to the stamens, in P. divaricata 
to both. The plants, especially the peduncles and calyces, are 
hairy to prevent creeping insects from ascending to the flowers. 
The general mechanism resembles in many points that of 
Hydrophyllum virginicum as described by Loew*, but in the 
latter the specialization of the corolline appendages has gone 
further, and they are turned into a tube for holding and 
protecting the honey. The series of species above described may 
also perhaps be compared with the series of species of Geranium, 
Epilobium, &c., as arranged by Miiller t, who shows how large 
and more conspicuous flowered species are best adapted for the 
cross, the inconspicuous for self-fertilization. 
The above described species cover only three of the seven 
subgenera. It is hoped to publish at some future time observa- 
tions on other species, as well as on some other genera of this 
hitherto almost untouched order. The flowers of Memophila 
and Hydrolea, besides other species of Phacelia (subgenus 
Cosmanthus), have been studied during the past year, but the 
results are not yet complete enough for publication. 
Mownarpa. 
The only published observations on the biology of this flower 
are those of Errera and Gevaert t, and of H. Miiller §, only the 
former paper dealing in any way with the floral mechanism. As 
* “ Bliitenbiologische Beitrige, I.,” Prings. Jahrb. xxii. 1891, p. 445. 
+t ‘Fertilization of Flowers,’ Eng. ed. 1883, pp: 155, 264, &e. 
t “Sur la structure et les modes de fécondation des fleurs,” Bull. Soc. Roy. 
Bot. Belg. xvii. 1878, p. 38, 
§ Lic. p. 477. 
