PASSIFLOREJE OF ECUADOR AND NEW GRANADA. 35 
short soft pubescence. They may be referred to the form *' gossy- 
piifolia.” The same remarks apply to another specimen under 
the same number gathered near Viota, New Granada —(3) 
“Peripa, Ecuador, alt. 200 met., E. André n. 4412.” A form 
with nearly glabrous striate stems, relatively large 3-lobed 
leaves, the central lobe ovate-acuminate, much larger than the 
lateral ones, all thinly covered with rather short appressed pu- 
bescence; no flowers or fruit; n. 4710, “in Peruvia, prope 
Payta” represents apparently the same form.—(4) “Daule, 
Ecuador, Ed. André n. 4142 (bis)." A form very similar to the 
preceding, but with the stems covered with long spreading hairs ; 
fruit yellowish olive, with a few coarse spreading hairs.—(5) 
* Las Juntas, Rio Dagua, Nov. Granata, alt. 300 met. Ed. André." 
Similar to the preceding, but rather less hairy, the fruits larger 
and almost entirely destitute of hairs.—(6) * Rio Quilcace, Nov. 
Granata, alt. 1388 met. ; flor. albid. vel pallide azurei. Ed. André 
n. 1626 ter." A form with densely hirsute pubescence without 
subjacent tomentum, with small, scarcely lobed leaves, and fruit 
with a few stiff hairs. According to M. André's drawing, the 
sepals are of a light bluish colour, the projecting ribs on the outer 
side green, the petals shorter than the sepals, also light blue; 
coronal threads blue, white at the base, smaller ones violet ; ovary 
downy, stigmas green. 
The forms of P. fætida, like those of P. suberosa, are so nume- 
rous and pass one into the other by so many gradations as to be 
almost inextricable as herbarium specimens. It is probable that 
in a fresh state it would be more easy to disentangle the forms 
and to correlate the different degrees of pubescence &c. with 
the local conditions under which the plant grows. MM. Triana 
and Planchon prefer to regard all or most of the forms as 
distinct species ; but, considering that the distinctive characteris- 
ties are adaptive rather than congenital or phyletic (see Trans. 
Linn. Soc. vol. xxvii. p. 621), it seems preferable to regard 
them as modifications of one very variable type. The character 
relied on by MM. Triana and Planchon to separate P. fetida 
(^ fruit hérissé de poils ") from P. hispida, DC. (* fruit glabre ”), 
does not, in my experience, deserve so much consideration as the 
authors cited give it. 
7. P. ArNIFOLIA, H. B. K. Nov. Gen. et Sp. ii. p. 136 ; Mast. 
in Flor. Brasil. l. c. p. 549, pro parte; Triana & Planch. l. c. 
p. 165. 
