that the flowers would be frequently visited by insects, and a little observation soon shows this to be a fact. On a warm sunny day it is almost impossible to watch a bed of this Orchid for any length of time without seeing numerous Diptera flitting from flower to flower, busily engaged in robbing them of their sweets.

If we now call to mind the manner in which the column arches over the lip, we can easily see that an insect crawling into the flower to get at the supply of nectar can hardly avoid touching one of the points of the rostellum, ranging almost directly over it; if it did so, the delicate exterior membrane would be at once ruptured, and the viscid mass firmly glued to the insect's back. Thus, on withdrawing from the flower, the visitor would carry away with it not only the portion of the rostellum which it had touched, but also the attached pair of pollinia. These (from each pollinium being nearly subdivided into two) would form four little projections standing rigidly erect on the back of the insect; and consequently, when conveyed to another flower, can hardly fail to strike the overhanging stigmatic chamber, which is sufficiently viscid to detach a portion, at least, of the pollinia from the body of the insect, thus ensuring the fertilisation of the flower.

As I have several times seen insects remove the pollinia, and on one occasion also seen a pollen-mass left on the stigma, there can be little doubt that fertilisation is conducted on this plan. That insect aid is absolutely required is proved by the fact that the pollinia remain in their cells, and never reach the stigma, when the plant is covered up or allowed to expand its flowers in a room. But, under natural conditions, the flowers are so frequently visited that the pollinia are generally removed directly after the opening of the blossoms; while the large proportion of capsules produced is good evidence of the completeness with which the visitors perform their duties. Out of eighty-seven flowers, borne on fourteen plants, no less than seventy-one matured capsules, and of those that had failed to do so, many were imperfect ones situated at the summit of the panicle, and probably incapable of producing seed. Another set, from a different locality, had borne forty-seven flowers, of which no less than forty-four had ripened capsules.

The fact that almost every perfect flower produces a capsule, is in remarkable contrast to what occurs in several other genera of our Orchids. For instance, Pterostylis is fertilised on a plan much more complex, and the co-adaptation of the various parts of the flower is so complete that almost every insect that fairly enters the flower must remove the pollinia, which is not the case in Acianthus. Yet, from some reason, probably from the want of sufficient attraction, the flowers are comparatively seldom visited, and consequently few capsules produced. In my account of the fertilisation of this genus, (Transactions of the N. Z. Institute, Vol. V, p. 356.) I have