which was found by Sir James Hector near Waitapu, the boat-harbour of the Takaka Valley in Massacre Bay.

The distribution of these shells, so far as it is at present known, is interesting. Busbyi appears to be coincident in range with the kauri forest, for I am not aware that any specimen has been found south of the parallel to the north of which that timber is now growing. The typical hochstetteri, of which the Waitapu specimen of Sir James Hector is an example, was found, as already stated, on the mountain-range between the valley of the Pelorus and that of the Maitai, which flows into the Nelson Haven. I have already mentioned the localities in which the so-called dark hochstetteri has been got, and it becomes an interesting point to determine how this variety should have a place on both sides of Cook Strait, and upon an island so isolated as Stephens Island, and yet should never have been found within the area from which the Picton shells have been obtained. I am not at present in a position to solve the puzzle, but I have thought that it might interest the members of the Society to hear some of the known facts in regard to the distribution of land-shells, in order, in the first place, that they may understand the difficulty of solving puzzles of a similar nature, and, in the next place, in the hope that it may stimulate observation and inquiry in relation to our own particular instances.

I need scarcely call your attention to the fact that questions affecting the distribution and dispersal on the face of our globe of all forms of life (using the words in their widest sense) have acquired extraordinary importance amongst naturalists since the publication of “The Origin of Species,” and that numberless theories have been propounded to account for the anomalies often presented to us in connection with this subject. One favourite theory, not yet entirely abandoned, has been founded upon supposed frequent and radical changes in the distribution of land and water, but this theory has gradually lost the support of those whose opportunities of observation have been most extensive, and we are being taught to look to the existence of other and yet of necessarily highly effective means of dispersal, in order to account for the facts which observation has brought under our notice.

It has been shown, however, that in considering the questions of distribution and dispersal no group of animals presents greater puzzles than land-shells, for, as Darwin pointed out, almost all oceanic islands, even the smallest and most isolated, are inhabited by members of this group, generally endemic, though frequently mixed with species found elsewhere; and yet it seems almost impossible to conceive how (except where the agency of man can be referred to) organisms which are very speedily killed by immersion in

salt water could have passed over the vast tracts of ocean which frequently separate their present habitats from the nearest continental areas in which closely-allied forms are found. Nevertheless, the fact that such islands as New Zealand, the Sandwich Group, the Canaries, the Cape de Verde Islands, St. Helena, and others equally remote from continental lands, possess numbers of endemic forms shows that the means of dispersal, whatever these may have been, were sufficient to bring about their occupation by a large variety of forms in this particular group of animals.

In connection with these facts attention has been pointedly called to the circumstance that the forms of land-shells are very stable, and that they have, indeed, continued with little change through many geological periods, and therefore that time must be treated as a most important factor in bringing about the distribution which we now find to exist. As an instance of this stability of form, Wallace has pointed but that most of the land-shells found in the Miocene and Pliocene formations in England are similar to, and many of them almost identical with, species still living in England; that, in the Eocene, ordinary forms of many of the existing genera, especially of Helix, Clausilia, Pupa, Bulimus, Glandina, Cyclostoma, Planorbis, Paludina, and Limnœa, have been met with, and he specially mentions that one species of Helix found in the British Eocene is still living in Texas, whilst some of the existing Brazilian sub-genera have been found in a fossil condition in Eocene strata in the south of France; and he further mentions, as a still greater proof of the stability of this group, the recent discovery, in the coal-measures of Nova Scotia, of two species of Helicidœ, both of living genera, some specimens of which were obtained from the hollow trunk of a Sigillaria, and others, in large quantities, in a bed full of stigmarian rootlets, and he adds that between the shells so found and living species of the same genera the most minute examination has failed to detect the slightest difference in form or microscopic structure. Looking to these facts he points out that, in estimating the importance of peculiarities or anomalies in the continued existence and geographical distribution of the present living land-shells, as compared with that of many of the higher forms of life, we must always bear in mind the possible and even probable higher antiquity of the former.

I have already mentioned all that is at present known of the local distribution of the specimens of Paryphanta now before you, but this peculiarity is merely analogous to that of some of the other forms of life, both animal and vegetable, in these Islands. Amongst our birds, for example, you will at once call to mind the restricted ranges of the huia, of Nestor notabilis, and of some other species. In our plant-life it

is well known that slight differences in latitude often determine the range of many species, and therefore the comparatively restricted range of the different shells before you need cause no surprise, more especially when we compare it with the peculiarities in this respect which are found to exist in other oceanic islands. The Sandwich Islands afford a notable example of such peculiarity. They have yielded between three and four hundred species of land-shells, all or nearly all of which are said to be endemic; and the average range of any species (according to the observations of the Rev. Mr. Gulick, the chief authority on this subject) is five or six square miles at the outside, whilst some are restricted to but one or two square miles, and only very few have the range of a whole island. It is even said that some valleys, and often each side of the same valley, possess their own peculiar species.

The case of Madeira is a somewhat singular one. Wallace tells us that about fifty-two species of land-shells have been found in Madeira and forty-two in the small adjacent island of Porto Santo, but that only twelve are common to both islands, though all, or almost all, are distinct from their nearest allies in Europe and North Africa.

Now, all this is very singular; and yet so great has been the stimulus given to inquiry by the writings of Darwin that we are gradually becoming acquainted, through the observations of a host of naturalists in all parts of the globe, with means of dispersal of this class of animals which must have acted for all time, and may fairly account for its present condition of distribution. Darwin himself made many experiments on the power of land-shells to resist sea-water, and these were sufficient to show that it was quite possible for them to be carried in driftwood for many hundred miles across the sea, and this, according to Wallace, coupled with their power of living for long periods without food, is, as I believe, probably one of the most effectual modes of their dispersal. As an instance of such means I may quote the following passage from Mr. Hickson's “Naturalist in North Celebes.” Speaking of the fact that many of the squirrels, rats, and bats of the Philippine Islands and of Celebes are identical, or very similar, and that this is not to be wondered at, for that they could easily be drifted from the one island to the other on drifting timber, he says that “the opportunities afforded to arboreal animals to emigrate from one island to another are not so infrequent as might be supposed. During the heavy rains of 1882 the Wanada River brought down vast numbers of forest-trees, and many of these must have drifted to sea with a considerable crew of squirrels, mice, caterpillars, and other animals.” And he mentions the remarkable fact that after the eruption of Krakatoa, in 1883, a female green