Let us consider the flower simply as a device to insure its own cross-fertilization. The insect is welcomed; it must alight and sip the nectar; in departing it must bear away this pollen upon its body, and convey it to the _next_ Arethusa blossom which it visits, and leave it upon its stigma. These are the conditions expressed; and how admirably they are fulfilled we may observe when we examine flower after flower of a group, and find their nectaries drained, their anther cells empty, and pollen upon all their stigmas. The nectar is here secreted in a well--not very deep--and the depth of this nectar from the entrance is of great significance among all the flowers, having distinct reference to the length of the tongue which is expected to sip it. In the Arethusa, it is true, the butterfly or moth might sip at the throat of the flower, but the long tongues of these insects might permit the nectary to be drained without bringing their bodies in contact with the stigma. Smaller insects might creep into the nectary and sip without the intended fulfilment. It is clear that to neither of such visitors is the welcome extended. What, then, are the conditions embodied? The insect must have a tongue of such a length that, when in the act of sipping, its head must pass beyond the anther well into the opening of the flower. Its body must be sufficiently large to come in contact with the anther. Such requisites are perfectly fulfilled by the humblebee, and we may well hazard the prophecy that the Bombus is the welcomed affinity of the flower. [Illustration: Fig. 4] The diagrams (Fig. 4) sufficiently illustrate the efficacy of the beautiful plan involved. At A the bee is seen sipping the nectar. His forward movement thus far to this point has only seemed to press the edge of the anther inward, and thus keep it even more effectually closed. As the bee retires (B), the backward motion opens the lid, and the sticky pollen is thus brought against the insect's back, where it adheres in a solid mass. He now flies to the next Arethusa blossom, enters it as before, and in retiring slides his back against the receptive viscid stigma, which retains a portion of the pollen, and thus effects the cross-fertilization (C). Professor Gray surmised that the pollen was withdrawn on the insect's head, and it might be so withdrawn, but in other allied orchids of the tribe Arethusae, however, in which the structure is very similar, the pollen is deposited on the thorax, and such is probably the fact in this species. In either case cross-fertilization would be effected. Nothing else is possible in the flower, and whether it is Bombus or not that effects it, the method is sufficiently evident.