addition to the thread-cell would seem to be essential to the urticating faculty," he says, when speaking of the jelly-fish, since these cells are present in species and parts that do not sting." It is to be regretted that he was not moved by this doubt to a closer examination of the evidence on which the urticating faculty rested; he would assuredly have been led to the belief that no superaddition to the thread-cell will account for the phenomenon.

But I waive the argument derived from such a source, and, confining myself to the anemones, ask the reader what he thinks of this awkward fact, namely, that these urticating cells are most abundant in parts which do not urticate? Only the tentacles have this power, and although they have numerous cells, the urtication cannot well be attributed to them, since these cells are more abundant in the convoluted bands, in the lining walls of the stomach, and in the blue spots which surround the oral disc in the smooth anemone - these spots, indeed, being made up of such cells and small granules-yet in not one of these parts can the slightest urtication be traced! How is this? If these cells are the nettling organs, why do they not nettle in those parts where they are most abundant? No one has thought of asking this question.

It thus appears that animals having the cells, have none of the power attributed to the cells; and that even in those animals which have the power, it is only present in the tentacles, where the cells are much less abundant than in parts not manifestng the power: the conclusion, therefore, presses on us that the power does not depend upon these cells. And this conclusion is strengthened every step we take. Thus the Anthea is of all anemones the most powerfully urticating; yet, if we compare its cells with those of other anemones, wo find them greatly inferior in quantity to those of the Daisy and Dianthus, and much inferior in size to those of Crassicornis, as well as less easily made to uncoil their threads. It has not been remarked, that whereas according to theory the thread should dart out almost instan

taneously on the slightest pressure; in point of fact it frequently cannot be pressed out at all, even when the whole force of the finger is exerted on the two pieces of glass between which it lies. From the very capricious way in which the threads dart out while under the microscope, and not under pressure, and from the frequent impossibility of pressing them out, I suspect that pressure has really nothing normally to do with the ejection of the thread.

Hitherto we have merely considered facts of observation; we shall now see them confirmed by experiment. Mr Gosse proposes to establish a new genus, named Sagartia, on this purely hypothetical function; including in it all those anemones which, like the Daisy and Dianthus, possess an abundance of peculiar white filaments, visible to the naked eye, which are protruded from the pores of the body and the mouth, when the animal is roughly handled. These filaments are seen, on examination, to be chiefly composed of the "urticating cells." Mr Gosse names the genus Sagartia, because Herodotus says of the Sagartians, that "when they engage with the enemy they throw out ropes which have nooses at the end, and whatever any one catches he drags towards himself, and they that are entangled in the coils are put to death.' The name, you perceive, is aptly chosen, that is, it would be, if the hypothesis of the filaments were not a figment. The filaments have no such lasso-like and murderous power. This Mr Gosse would deny; and I remember he somewhere records an observation which would perhaps quite satisfy him that his denial has good ground to stand on. He relates that he once saw a small fish in the convulsions of agony, with one of these filaments in its mouth; it shortly expired, and he unhesitatingly concludes from this fact that the Sagartia "will attack even vertebrate animals." It is a matter of surprise and regret that Mr Gosse, having once made such an observation, did not feel the imperative necessity of repeating and varying the fact, so as to be sure that the death was not a mere coincidence. If the filament had the power which this single observation fairly

seemed to suggest, nothing could be easier than to establish the fact by experiment. But, I repeat, no one has seen the necessity for the verification of an hypothesis so plausible; and Mr Gosse, like all his predecessors, was content with recording his observation, as if it carried the point. Not being so content, I tested it thus: After irritating a dianthus till it sent out a great many filaments, I dropped a very tiny annelid among them, and entangled it completely in their meshes. Yet lo! these filaments, which are said to possess so powerful a faculty of urtication that even vertebrate animals are killed by them, had no other effect upon a soft annelid than that of detaining it in their meshes, from which it shortly freed itself and wriggled away unhurt. Nor was I yet satisfied; placing a tiny crustacean, of the shrimp family, among the filaments of another dianthus, I saw it remain there enveloped, but apparently quite comfortable, not in the least so desirous of escaping as one would expect if it were being "nettled" all over; and when I lurched the jar it swam away. I have since repeated this experiment with entomostraca and annelids, without once detecting the slightest indication of their being more incommoded by the filaments than they would have been by threads of silk. Mr Gosse, indeed, not only maintains that these filaments are weapons of offence, but he actually goes so far as to suggest that the blue spherules which surround the disc of the Mesembryanthemum may "represent the function of these missile filaments" because they are composed of the thread capsules. But I repeat, the hypothesis which assigns to the thread capsules a function of urtication or prehension, is an hypothesis without a single fact to warrant it, and is contradicted by the various facts I have just adduced. Ehrenberg has very unwarrantably given an ideal figure of a hydra in the act of seizing its prey, with the hooks of the thread-cell extended; but, as Siebold truly remarks, the animal is never seen thus; and I will add that it is always seen in precisely the

contrary aspect, namely, the blunt end of the cell being in contact with the animal, the hook and thread being turned towards the polype.

I have reserved one fact as the coupde-grace. Having shown that the parts most abundantly supplied with these "urticating cells" do not urticate, I can now remove the last vestige of doubt by the fact that the cell itself from the tentacle of an anemone, when seen to eject the thread and touch an animalcule, does not kill or disable that animalcule a fact I witnessed when examining the cells under the microscope. This not only gives the coup-de-grace to the general hypothesis, but even sets aside that suggestion of Professor Owen's respecting the probable superaddition to the urticating cell which is to distinguish it from cells in those parts destitute of the power.

*

The foregoing discussion has had a purpose beyond that of rectifying an universal errorr-the purpose of pointing a lesson in comparative anatomy. The greatest living_experimental physiologist, Claude Bernard, has recently insisted with emphasis on the importance of recognising "anatomical deduction' to be a fruitful source of error. He warns us against attempting to deduce a function from mere inspection of the organ, without seeing that organ in operation, and applying to it the test of experiment. As a case of pure deduction, this hypothesis of the "urticating cells" seemed to command, and did command, instantaneous assent; but on submitting it to verification, we find the hypothesis to be an error. To the philosophical mind, therefore, there will have been an interest in the foregoing discussion greater than any interest issuing out of the mere conclusion respecting the thread-capsules.

There are other new facts which

were yielded to patient investigation, but, having limits necessarily somewhat circumscribed, this Magazine cannot contain all facts, even were its readers of unappeasable appetite; so I will confine myself to the single discovery of the reproductive system in the anemones, that being of some

more complex organisms perform each of these functions by a special apparatus of organs, yet these organs themselves are originally developed from the envelope. We may, ideally, reduce even a mammal to a cylindrical envelope folded inwards at each end; from the enfolded skin are developed all nutritive and reproductive organs, while the nervous system and its osseous sheath are developed in the space between the outer and inner walls of the envelope. Thus every advance in complexity of organisation takes place through a gradual differentiation, or specialisation, of the general envelope. These important synonyms, differentiation and specialisation, I will explain by illustrating the law to which they point, namely, the law of animal development first enunciated by Goethe, and strikingly applied by Von Baer: Development is always from the general to the particular, from the homogeneous to the heterogeneous, from the simple to the complex; and this by a gradual series of differentiations. When we say an organ has been formed out of a tissue, we say a differentiation has taken place; and the function, e. g. respiration, which before was performed by the general tissue, is now specialised, i. e. performed by that special organ. A homogeneous mass of organic matter, such as the Amoeba, which has no organ whatever, performs all the functions of assimilation, respiration, progression and reproduction, by its general mass, not by any special organs. The process of differentiation by which special organs are gradually developed in the ascending scale of the animal series, is equally exhibited in any particular case of development. Thus if we follow the formation of the human hand, we find first a differentiation between the carpus or wrist, and the metacarpus or hand; next the fingers are differentiated, but, without any division into separate segments this takes place later; then we have a separation between the soft and hard parts, the cartilage separating from the plastic mass; then these cartilages become osseous; and in the soft plastic mass we distinguish differentiations into muscle, tendon, skin, &c.; when the single

tissues are thus separated we may begin to trace differentiations in the skin, such as the papillæ, the secreting glands, and so forth: till, from a homogeneous mass of cells, we have traced the development of that marvellous and complex structure, the human hand.

Applying this torch to the obscure question of the reproductive system of the anemones, it at once discloses to us that the anemone, being of a very simple organisation, almost entirely envelope, we shall be wrong if we expect to find in it a high complexity of special organs. Anatomists, indeed, have often neglected such a consideration, and have worried themselves in the search after organs, which a priori we may decide were not likely to be present. They have sought for and "discovered" nerves and ganglia, each discoverer scornfully rejecting the alleged discovery of his predecessor, and declaring the nerves were in a totally different locality, while no one anatomist could find them anywhere after another. They have worried themselves about the respiration of the anemone, not perceiving that respiration, like circulation and other functions elsewhere dependent on a special apparatus, was here performed in a direct and general manner. They have not suspected that reproduction takes place in the anemone, much in the same way as in the freshwater polype-not in any special and permanent apparatus of organs, such as ovary, oviduct, &c., but by a temporary specialisation of the general envelope including an accumulation of germ-cells and sperm-cells. I am aware that special organs called ovaries are described in all books, and that some writers describe an oviduct-which only exists in their imagination, for no duct of any kind is found. Of course, no philosophical a priori conclusion could be permitted to stand up in contradiction to observed fact; if the organs are there, it is of no use deductively establishing their non-existence. But are they there ?

When I first commenced the investigation of anemones, I had no reason whatever to doubt the statement so generally and confidently

made, that the convoluted bands were the organs in question. At the end of the first week my doubts began. These convoluted bands contained no trace of ova, but instead thereof they contained vast quantities of those thread capsules which I then believed to be urticating cells. This was the last place in the world where one might expect to find of fensive weapons; and misled by the belief in these cells, I was led to question the function of the convoluted bands. Questioning, of course, meant something more than supine doubt. I began on the 14th May to examine closely into the evidence, and on the 12th June I was fortunate enough to confirm all doubts by the discovery of the real ovaries (such as they are) in a large Crassicornis: here there were no thread capsules, but abundance of unmistakable ova, each with its "vesicle of Purkinje." The thrill of delight with which the assurance broke upon me may be conceived. After exploring several other anemones, to remove all lingering doubt, I hastened to communicate the discovery to my friend Mr Tugwell, in whose presence I again displayed the organs. At that time I, of course, believed that the grapelike cluster in which the ova were lying, were true and permanent ovaries; but having since been frequently unable to detect them in adult specimens, and never in young specimens, I come to the conclusion that these ovaries are temporary organs, formed by an accumulation of germ-cells in various parts of the lining membrane of the envelope; that, in fact, they represent the first rudimentary state of what in higher animals becomes the special organ. This conclusion is, however, purely theoretical, and I will now state what any one may see, who examines an adult fresh from the rockpool or tank. With a rapid but not deep incision we lay open the envelope from the outside; the convoluted bands will bulge through the opening; but if we are vigilant and brush these aside, we shall perceive certain lobular or grapelike masses of darker colour, almost entirely hidden by these bands, but growing from the walls of the envelope. They are not situated

I

in any precise spot; near the base, about the centre, and close to the disc, they may be found: nor are they in every interseptal space; sometimes we may make three or four incisions before detecting them. Once seen, they will easily be distinguished from the convoluted bands, although so difficult is it to remove them without at the same time removing some of the bands, that to this cause alone can I attribute the long continuance of the opinion that the bands were the true ovaries. For it should be observed that several writers have discovered the ova, and one at least (Spix) seems to have seen the ovaries; but that no one had clearly discriminated and described the organs, is evident in the confusion which our text-books exhibit on the topic. believe I may not only claim the discovery, as having been made independently, and without any knowledge of what Spix had seen, but also as having for the first time discriminated both anatomically and physiologically the ovaries from the convoluted bands, so as to clear up all confusion. I am not even certain that Spix recognised the real organs, since he describes ducts opening into the stomach by several apertures, when in truth there are no ducts, and the aperture at the base of the stomach is one, not several. It is from Dr Johnston's History of British Zoophytes (for a hasty reading of which I was indebted to Mr Tugwell, after I had made the discovery) that I gather what Spix said. He describes the ovaries" as forming several grapelike clusters situated in the interseptal spaces with ducts which open into the base of the stomach by several apertures, and hence the ova are presumed to gain their freedom by traversing the stomach and mouth. De Blainville doubts this, being led to believe it more probable that the oviducts may open into the labial rim as they do in the asteroid polypes." From this it appears that even if Spix detected the ovaries, he did not accurately discriminate them from the convoluted bands; he did not accurately describe their structure, for he speaks of ducts where no ducts exist; he did not understand their nature, as temporary specialisa