The successful issue of the observations is so certain that they can be best described in the form of directions to a person who should be desirous of seeing them for himself, followed by the prediction of what will be observed by him.

§ I. Two vessels of salt solution are to be prepared, and connected with the capillary electrometer by electrodes. The various extremities of the observer are to be dipped into the salt solution, while the capillary column is watched. Electrical variations, apparently synchronous with the heart's pulse, will be observed with certain combinations rather than with others, and the results (on a normal person with the heart pointing to the left) will be as follows:

produced if at any time the apex and base of the ventricles differ in potential. oo is the line of zero potential at right angles to C C.

a a a are equipotential lines round a supposed focus A. bbb are equipotential lines round a supposed focus B. Any lead off from two superficial points a a or bb is unfavourable. Any lead off from two points a b is favourable to the manifestation of electromotive differences originating at the heart. This will have been demonstrated by the experiments directed to be made.

§ II. On a quadruped (dog, cat, rabbit) the results will come out somewhat differently. The heart occupies an approximately median position, so that the asymmetry observed on man does not hold good with the abovenamed animals. In these the current axis will be along a median longitudinal line; the line of zero potential will be at right angles to it, i.e. transverse.

This can be verified by trial with very little trouble. A quadruped is led off by the various extremities and orifices immediately after death before the heart has ceased to beat; or a dog may be trained to stand quiet with his feet in dishes of salt solution (I have a large and well-disposed dog who will stand thus by the hour. However the test be made, the results will come out as follows:-

§ III. Upon these two proofs may be piled a third proof of the correctness of the facts and of their explanation. Cases of situs viscerum inversus are to be found; the viscera of such people are situated as those of a normal person seen in a mirror; i.e. inter alia, the heart points to the right. I have examined two such cases, with results exactly as anticipated, viz. the favourable combinations, 4, 5, and 7, of a normal subject (§ I.) are unfavourable in the case of situs inversus, while the unfavourable combinations, 2, 3, and 8, are favourable. Combinations 1, 9, and 10 are favourable, and 6 is unfavourable in both cases, there being the notable peculiarity as regards 1 that the variations are reversed in direction in each of the two cases. The significance of this point will be obvious to the reader who has followed the facts up to this point: in both cases we have a favourable combination, but a reversal of points a and b.

§ IV. As regards the character and direction of each cardiac variation, it will be found to be composed of two phases, the first short, sharp, and difficult to read as regards direction, the second comparatively prolonged and easy to read. The second phase clearly indicates negativity of the heart's base, the first phase less clearly negativity of the heart's apex-facts which testify that the contraction begins at the apex and ends at the base of the ventricles. The auricular contraction does not affect any electrometer I have used.

"Paw" is used as an abbreviation for anterior extremity; "foot" for

posterior extremity.

If I may venture to forecast the manner in which these statements may receive from independent sources that verification which any statement requires before it can be accepted as a correct representation of fact, I should say that as regards § I. no contradiction will arise unless the first case tested should happen to be that of a person with the heart occupying an unusually median position, when the favourable and unfavourable cases, though still distinguishable, may be less so than if the heart occupied its usual oblique position pointing to the left. In any case, however, the variation will be found more marked with a favourable than an unfavourable combination. As regards § II., the statements

made can be verified as soon as tested upon a recently killed cat or upon a properly educated dog. The verification of § III. only requires that a suitable case should be discovered. As regards the character of the variation, it is probable that its diphasic character may be overlooked at the first glance, but (in a favourable case) this character will soon be apparent. As regards direction, that of the second phase will be determined without much difficulty, but that of the first will be found very difficult to seize. I was not able to make up my mind about it until I had obtained successful photographs of the movements on a quick-travelling sensitive plate. AUGUSTUS D. WALLER.

I

THE MAXIMUM OF MIRA CETI.

AM anxious to call the attention of observers to the present spectrum of Mira, which arrived at its maximum brilliancy on the 15th inst. I pointed out recently (NATURE, May 24, p. 79) that stars of the group to which Mira belongs are sparse meteorite-swarms like comets, and that, when variable, the variability is produced by collisions between two swarms, the centres of which are nearest together (periastron passage) at maximum. Broadly speaking, then, we may regard variables of this class as incipient double stars, or condensing swarms with double nuclei, the invisibility of the companion being due to its nearness to the primary, or to its

faintness. It is obvious that variability will occur mostly in the swarms having a mean condensation, for the reason that at first the meteorites are too far apart for many collisions to occur, and that, finally, the outliers of the major swarm are drawn within the orbit of the smaller revolving one, so that it passes clear.

The present maximum of Mira tests my hypothesis, and its brightness is such that a small telescope and a Maclean's spectroscopic eye-piece are all that are necessary to see in how striking a manner the test is borne. The two brightest bands now visible are at A 517 and A 546, precisely where these are seen in the brightest comets. The former is the brightest carbon fluting seen in the spectrum of the Bunsen flame, or spirit-lamp,

and the other, at 546, is the citron carbon fluting beginning at 564, but modified by the masking effects of the manganese absorption fluting at 558, and also that of lead at 546.

The blackness of the spaces between the bright flutings shows that there can be very little continuous spectrum from the meteorites, and therefore that the absorption is that of the light of the carbon flutings.

The mean spectrum of Mira is that of a star like B Pegasi, which I have shown to consist of bright carbon flutings, and dark flutings of magnesium, manganese, iron, lead, and barium. In B Pegasi, as in Mira under mean conditions, the carbon is somewhat faint, but in a Herculis it is very bright. The general effect of the conditions of maximum of Mira therefore seems to be

FLORA OF THE KERMADEC ISLANDS.

UPWARDS of thirty years ago Sir Joseph Hooker published an account of the botany of Raoul or Sunday Island, one of the Kermadec Group (Journal of the Linnean Society, i. pp. 125-29), founded upon a small collection made by McGillivray and Milne, naturalists attached to H.M.S. Herald. This collection consisted of forty-two species, of which twenty were flowering plants, and the rest ferns and lycopods and the most interesting circumstance connected with it was "the identity of most of the flowering plants, and all but one of the ferns, with those of New Zealand.”

:

In 1885, Mr. J. T. Arundel presented to the Kew Herbarium a collection of fourteen species from Meyer, a small rocky islet about a mile and a half north of Sunday Island. Poor as it was, it contained half-a-dozen plants not previously known from the group, though they are all included in the collection referred to below.

Since then, no further light has been thrown on this insular flora, until the quite recent appearance (Transactions of the New Zealand Institute, xx. pp. 151-81) of a paper by Mr. T. F. Cheeseman, Curator of the Auckland Museum, New Zealand, a copy of which was kindly forwarded to the writer. Mr. Cheeseman was permitted, through the kind offices of Mr. Percy Smith, the Assistant Surveyor-General of New Zealand, to accompany the expedition despatched last year for the purpose of formally annexing the group to the colony of New Zealand. If Mr. Cheeseman has not succeeded in exhausting the botany of the Kermadec Islands, which, of course, is hardly probable, the undiscovered species cannot materially affect the question of the origin of the vegetation. But before giving the results of his investigations, it will be useful to indicate the position and extent of the islands.

There are four islands lying at great distances apart, between 29° 10' and 31° 30' S. lat., and stretching in a southwest and a north-east direction, like New Zealand itself, the nearest point of which is between 500 and 600 miles distant. Raoul or Sunday Island is the largest and the farthest from New Zealand, being twenty miles in circumference, and about 640 miles from Auckland, and a little less than that distance from Tonga. Macaulay, the next in size, is sixty-eight miles to the south-west of Sunday Island; and Curtis and L'Espérance, still farther to the south-west, are little more than rocks. The expedition failed to land on the last-named island, and the visit to Curtis Island was of very brief duration, hence the botany relates almost exclusively to Sunday and Macaulay Islands.

The group is of volcanic origin, and the greatest elevation in Sunday Island is 1720 feet, while Macaulay nowhere reaches quite half that height.

Altogether Mr. Cheeseman collected 115 indigenous vascular plants, eighty-four being phanerogams and thirty-one cryptogams, and only five of them were regarded as endemic. In addition to the foregoing, twenty-six species of naturalized plants, chiefly European weeds, were observed or collected.

Of the 115 indigenous species, no fewer than eighty-five are also found in New Zealand, though only fourteen of these are absolutely confined to the two localities. Fortyfour species are found in Norfolk Island, forty of which also occur in New Zealand, and only two are apparently confined to Norfolk Island and the Kermadecs. Forty species extend to Lord Howe's Island, but thirty-four of these are also in New Zealand, and none of the peculiar plants of Lord Howe's Island reach the Kermadecs. Seventy-six of the species are common to Australia, sixtythree of them being also in New Zealand, and none of them otherwise peculiar to Australia. Lastly, forty-seven are found in Polynesia, and thirty-one of these also occur in New Zealand.

The foregoing data, as Mr. Cheeseman observes, point unmistakably to New Zealand as the source of the greater part of the flora of the Kermadec Islands. How the plants reached these islands is an interesting question. Mr. Cheeseman is prepared to admit a former northwestern extension of New Zealand; but, after a careful examination of the evidence, he arrives at the conclusion that the Kermadec Islands have always been isolated, or, at least, have not formed part of any other land since the Secondary period. Spores of the ferns may have been conveyed by winds; and ocean currents and birds, it may well be conceived, have operated in stocking the islands with flowering plants. Most of the birds are New Zealand species, and the presence of Kauri logs, of different dates and brands, stranded on various parts of the beach, is convincing evidence of the direction of ocean currents. Moreover, the composition of the flora strongly supports this theory.

Sunday Island is the only one of the group on which there is anything approaching arboreous vegetation, and this, with the exception of a small area of the crater, is clothed with forest from the sea-shore to the tops of the highest peaks. The prevailing tree is Metrosideros polymorpha, one of the most characteristic trees of Polynesia, especially of the smaller islands, reaching the Sandwich, Marquesas, and Pitcairn Islands; but this particular species does not occur in New Zealand nor in Australia.

Next to the Metrosideros in abundance and con

spicuousness is a palm, which Mr. Cheeseman thinks may be identical with the Norfolk Island Rhopalostylis Baueri (Areca Baueri). In some places this grows gregariously, forming large groves.

Ferns are everywhere abundant, varied, and luxuriant ; and the endemic tree-fern, Cyathea Milnei, is very plentiful, and handsome withal, rising to a height of 50 or 60 feet. Prominent among the New Zealand trees are Corynocarpus lævigatus, Myoporum lætum, Melicope ternata, Melicytus ramiflorus, and Panax arboreum. Cordyline terminale, the widely-spread Polynesian “ Ti,” and Pisonia Brunoniana, Pittosporum crassifolium, Coprosma acutifolia, and C. petiolata, natives of New Zealand, are other elements deserving of notice.

The herbaceous vegetation includes no plants with very conspicuous flowers, but there are two orchidsnamely, Acianthus Sinclairii, a native of New Zealand, and Microtis porrifolia, which also inhabits both New Zealand and Australia.

Macaulay Island was entirely covered with a beautiful sward of natural grass, supposed to be composed of a species of Poa and an Agrostis, but in the absence of flowers they were indeterminable.

Students of botanical geography will find much more that is interesting in Mr. Cheeseman's valuable paper, from which I have extracted the principal facts. W. BOTTING HEMSLEY.

Thou art a shock to my æsthetic sense, And offerest no kind of recompense In way of use; of every function shorn, Except to act as basis for a corn. When thou art gone I'll still maintain my grace, Still walk erect wherever I may be ; Still I'll belong to the athletic race,

Waltz with the fair, and kick mine enemy! So pace Schopenhauers, and pace Mallocks When I've acquired a hypertrophied hallux, To monodactyle type thus simplified, Life shall be simpler too, and so-beatified.

WE lately (Sept. 6, p. 437) printed an account of the formation of the Australasian Association for the Advancement of Science. If we may judge from the newspaper reports which have now reached this country, the first general meeting of the Association seems to have been remarkably successful. The session began at the Sydney University on Tuesday evening, August 28. Lord Carrington opened the proceedings with a short speech, and then an address was delivered by Mr. H. C. Russell, the President. On the following day the sectional meetings began, and their work went on during the remainder of the week. About 110 papers were sent in by students of various branches of science, and a considerable number of them will be published in full in the first volume soon to be issued by the Association. The members had an opportunity of taking part in several pleasant excursions, and much hospitality was shown to visitors by leading citizens. At the time of the meeting there were about 850 members, and it is confidently anticipated that next year this number will be largely increased. The next meeting is to be held in Melbourne, and Baron Sir Ferdinand von Müller, the Government Botanist of Victoria, is the President-elect. In 1890 the Association will meet in New Zealand.

THE following is the list of names to be submitted, at the annual meeting (November 8) of the London Mathematical Society, for the new Council:-For President, J. J. Walker, F.R.S.; for Vice-Presidents, Sir J. Cockle, F.R.S., E. B. Elliott, and Prof. Greenhill, F. R.S. The Treasurer and Hon. Secretaries remain unaltered. The other members are: A. B. Basset, Dr. Glaisher, F.R.S., Messrs. J. Hammond, H. Hart,

J. Larmor, C. Leudesdorf, and S. Roberts, F.R.S., Captain P. A. Macmahon, R. A., and Dr. Routh, F. R.S. It is proposed that the vacancies caused by the withdrawal of Lord Rayleigh, Sec. R. S., and the lamented recent death of Arthur Buchheim, shall be filled up by Messrs. Basset and Routh, as above.

H.M.S. Jackal, which has been engaged, under the direction of the Scientific Committee of the Scottish Fishery Board, in a cruise of physical investigation in the North Sea, recently returned to Granton. The course was along the east coast to the Orkney and Shetland Islands, and then to Bergen, Copenhagen, and Kiel. The physical work was carried on by Dr. Gibson, of the Chemistry Department of the Edinburgh University, assisted by Dr. Hunter Stewart and Mr. F. M. Gibson; and owing to the exceptionally favourable weather a large number of stations were formed at various parts of the route, at which series of temperature observations were taken, the density and alkalinity of the water determined, and samples preserved for analytical examination. Dr. Gibson had interviews with most of those conducting scientific fishery work in the countries visited, including Mr. Buch of Bergen, Dr. Paulsen, Lieut. Drechsel, Dr. Pettersen,

and Mr. Feddersen of Copenhagen, and Prof. Karsten of the Kiel Commission; and we understand these conferences may result in closer co-operation between the various countries, in regard to the method and scope of scientific fishery investigations.

THE members of the International Commission of Weights and Measures have finished their session at the Pavillon de Breteuil, Paris. The making of standard metres is progressing, and next year they will be distributed to the various Governments. The guarantee of the Bureau extends to the thousandth of a millimetre and the ten-thousandth of a gramme.

THERE are now on the books of the Institution of Civil Engineers 1614 members, 2499 associate members, 458 associates, 19 honorary members, and 939 students, together 5529, being an increase at the rate of 3 per cent. during the past twelve months.

A SPECIMEN of the sword-fish (Xphias) was captured some days ago in Long Reach, Milton Creek, Sittingbourne, by a bargeman. The fish measured 5 feet 2 inches from end of tail to tip of sword.

AN Agricultural and Industrial Exhibition was opened at Mysore by the Maharajah on the 17th inst.

AT a recent meeting of the Bombay Natural History Society, the idea of starting a Zoological Garden in that city was mooted by Mr. H. M. Phipson, the Honorary Secretary of the Society, and was warmly taken up. It was stated that the Society has been compelled to refuse large numbers of valuable specimens of animals offered to it. All that is asked from the Government is that they shall grant a site, and it is hoped that they may see their way to do so.

DR. J. C. Cox lately described, at a meeting of the Linnean Society of New South Wales, two very remarkable female figures, modelled in wax, obtained in an aboriginal camp at Miriam Vale, near the head of the Calliope River, Rockhampton. These figures are said to be the only examples of plastic art ever discovered among the Australian aboriginals.

In the Report of the Superintendent of the Adelaide Botanic Garden for the past year it is stated that the insect-powder plant (Fyrethrum cinerariafolium, Trevir.), roseum, and carneum, Bibrst.), and the cheesemaker (Withania coagulans, Dun.), which were introduced into the Garden a few years ago, have found a congenial climate there, and have prospered wherever they were planted in the colony. Eland's Boontges (Elephantorrhiza Burchellii, Benth.), which has also been recently introduced, does fairly well. In winter nothing remains of this plant but the roots, which contain tannic acid. A number of cuttings Almeria, have thriven wonderfully in the Garden. There are from the Daira grape, a valuable species which comes from now in the palm house 180 species and varieties of palms. The Museum of Economic Botany attached to the Garden has been enriched during the past year by 1795 articles, amongst the more remarkable of which was a collection sent by the Sultan of Johore, one of the specimens being a sample of sugar prepared

from the cocoa-nut.

STUDENTS of the Caucasian languages will be glad to learn that the second volume of Baron Uslar's work, "The EthnoIt congraphy of the Caucasus," has been published at Tiflis. tains his "Tchetchen Language," and, in an appendix, several articles on the epics of the Caucasian mountaineers, on the study of the Caucasian languages and their alphabets, as also a translation of Schiefner's "Tchetchensche Studien," and a collection of Tchetchen proverbs and tales about Nasr-eddin, by J. Bartolomei.

IN connection with the discussion on "Valency" at the Bath meeting of the British Association, referred to in last week's NATURE, Prof. Meldola read a paper on the constitution of the

azonaphthol compounds, in which he drew attention to the fact that the properties of these important colouring-matters could only be satisfactorily explained by admitting that they contained oxygen in the tetravalent condition.

THE vapour-densities of the chlorides of chromium have, for the first time, been determined by Profs. Nilson and Pettersson, of Stockholm. The interest attaching especially to the chromic chloride, hitherto known as Cr,Cl, in view of the recent redeterminations of the densities of the corresponding chlorides of aluminium and iron, gives more than secondary importance to the work of the Swedish chemists. Readers of NATURE will

remember that these recent experiments by the indefatigable workers just mentioned, and by Prof. Victor Meyer and his coworkers at Göttingen, upon the composition of the molecules of the chlorides of aluminium and iron, resulted in the conclusion that the double formulæ, Al,Cl, and Fe,Cl, must be abandoned in favour of the simpler formula, AICI, and FeCl3. This, of course, meant that our old notions as to the tetrad nature of these elements were incorrect, and that in reality they behave as triads. Profs. Nilson and Pettersson now clinch the matter by showing that chromium, which in many respects so much resembles aluminium and iron, behaves in precisely the same way. Chromic chloride was fortunately obtained in beautiful laminated crystals of almost perfect purity. The minute traces of absorbed moisture were readily eliminated by gently warming in a current of dry carbonic acid gas; when this was accomplished the requisite quantity was weighed out into a small platinum capsule in those experiments which were conducted in the platinum density apparatus, and in small pieces of ignited porous tubing when the porcelain apparatus was employed. The chloride was found to vaporize very slowly indeed at 1065° C., precluding the possibility of taking densities below that temperature; however, at this comparatively low temperature, the density was 6135. Now CrCl corresponds to a density of 5'478, while CrCl must of necessity require a number twice as great, and hence cannot exist in the gaseous state. On increasing the temperature to 1190°, the value of 5'517 was obtained, which remained practically constant up to nearly 1300°. Over 1300° the molecules of CrCl commence to break up into those of CrCl, and free chlorine. This is a most decisive result, and one which cannot possibly lead to any other conclusion than the adoption of the formula CrCl3. It is only fair to mention that Messrs. Friedel and Crafts on carrying out vapour-density determinations of aluminium chloride by Dumas's method for 250° above its boiling-point (183), have very recently (obtained results which appear to indicate that this chloride may condense to the double molecule AlCl, at these comparatively low temperatures. How. ever this may be, there can be no doubt in the cases of iron and chromium that the triad formula is the only one compatible with experiment, and we shall be very glad to see the doubt in case of aluminium completely cleared up by further experiments. The determinations in the case of the lower chloride of chromium, CrCl,, have been made under great experimental difficulties. This substance is the most difficultly volatilized of any yet submitted to vapour-density determinations. It required the most intense heat of the hottest procurable furnace, and even then was only very slowly converted into vapour. It was obtained perfectly pure by reduction of the chromic chloride utilized for the former experiments, by gently heating in a stream of hydrogen. At the lowest observable temperature, 1300°-1400° C., the density was found to be 78, considerably lower than the number required by Cr,Cl. On further increasing the heat to 1600, the density gradually diminished to 6'2, showing that at some still higher temperature one would finally attain the value 4'25 corresponding to CrCl2. Hence chromous chloride again resembles ferrous chloride, the only difference being that the former is much more difficult to vaporize.

AN exceedingly useful and handy résumé of results in the "modern geometry of the triangle" is published in the jas issued Proceedings of the Association Française pour l'Avance ment des Sciences, Congrès de Toulouse, 1887. It is entitled "Premier Inventaire de la Géométrie du Triangle," by M. E Vigarié. A second " Inventaire," which the author proposes draw up, will be occupied with the extensions to certain Harmonic) quadrilaterals and polygons, and to space figures.

WE have received Part 3 of "A Catalogue of the Mot of India," compiled by E. C. Cotes, First Assistant to the Superintendent, Indian Museum, and Colonel C. Swinhoe. ( the first two parts, dealing respectively with Sphinges an Bombyces, we have already given some account (NATURE, vol. xxxvii. p. 386). The present part deals with Noctues Pseudo-Deltoides, and Deltoides.

THE Trustees of the Australian Museum, Sydney, have issued Part I. of a catalogue of the fishes in the collection of the Museum. It relates to recent palæichthyan fishes, and has been compiled by Mr. J. Douglas Ogilby.

A REMARKABLE book on "The Butterflies of the Eastern United States and Canada, with especial reference to New England," by S. H. Scudder, of Cambridge, Mass., U.S.A., is about to be published in monthly parts. It will be completed in twelve parts, the first of which will appear in November. The preparation of this elaborate work was first announced by the author in 1869. Since that time he has had it always in hand, and during the last eight years he has devoted to it undivided attention. According to the prospectus which has been issued, Mr. Scudder has not only availed himself of the personal aid of a host of willing friends and correspondents, who have confided to him their voluminous field notes and numerous specimens, but he has carefully gleaned every fact of value from the natural history journals and other publications, and supplemented all by his thirty-five years' experience in the field. It is claimed that no systematic work on butterflies has ever appeared in any language comparable with it in the complete elaboration of a single limited fauna, in attention to every stage of life, in thorough and excellent illustration of every period of the butterfly's existence, and in careful detail of all structural features. The book will contain seventeen plates of butterflies, six of eggs, eleven of caterpillars, two of the nests of caterpillars, three of chrysalides, two of parasites, thirty-three of :tructural details in all stages of life, nineteen maps and groups of maps to illustrate the geographical distribution of the butterflies, and three portraits of early naturalists of America-in all, about two thousand figures on ninety-six plates, of which forty or more will be coloured. The printing of the plates was begun three years ago, and is now nearly finished.

A THIRD edition of Mr. R. Milne Murray's "Chemical Notes and Equations (Maclachlan and Stewart, Edinburgh) has been issued. The book is intended for the use of students. In this edition a section on the electrolysis of salts has been introduced, and some additions have been made to the descriptive part of the work.

THE latest number (No. 3, vol. iii.) of the Journal of the Bombay Natural History Society contains, amongst other papers unscientific notes on the tiger, by J. D. Inverarity; butterflies and ants, by Lionel de Nicéville; on the Lepidoptera of Karachi and its neighbourhood (part 2), by Colonel Swinhoe; notes on some bees and wasps from Burmah, by Captain C. T. Bingham; notes on the origin of the belief in the bis-cobra, by G. A. Da Gama. Mr. Da Gama says that the term bis-cobra is not of Oriental origin, but is a contraction of the Portuguese bicho-de-cobra. The early Portuguese settlers in India named the animals they met with from their most prominent features.