tion must be larger than those of either of the two substances when separate, the solution consisting of compounds or hydrates of the two. I showed, moreover, in the paper above referred to that the hydrate theory of solutions was quite capable of accounting for and explaining the fact that the dissolved substance may for many purposes be regarded as being in a quasigaseous condition in weak solutions, and that calculations based on the idea of its being truly gaseous would yield very nearly correct results.

More

The hydrate theory will also, as I showed, give an explanation of the fact that electrolytes will give abnormally high osmotic pressures, and that the magnitude of these pressures can be calculated from their electric conductivity: and the explanation based on this theory also obviates many of the objections to which the idea of dissociation into ions is open. over, the only critical experiment which, as far as I know, has ever been made to test the validity of the dissociation hypothesis, gives an unequivocal answer against it, and in favour of the hydrate theory. When, for instance, sulphuric acid is dissolved in excess of water, it is represented by the dissociationists as splitting up into its ions, so that the solution will contain more acting units (ions and molecules) than the acid and water together contained before they were mixed whereas, on the hydrate theory, combination will have occurred, and there will be fewer acting units present. The number of acting units may be ascertained by ob. serving the depression produced by the solution on some other solvent, such as acetic acid (that is, by using the very method which the dissociationists use to prove the supposed dissociation of substances), and when this is done it is found that the sulphuric acid solution contains fewer, instead of more, units than the acid and water separately.

Even if the above were the only arguments to be urged, it is evident that although the idea of the dissolved substance being gaseous and often dissociated may be a good working hypothesis for the directing of investigation, it can scarcely be accepted as a true theory of the nature of solutions.

SPENCER PICKERING.

On a Supposed Law of Metazoan Development. IT is difficult not to feel disappointed that Dr. Beard has gi en only "a preliminary sketch by way of clearing the groundi (NATURE, vol. xlvii. p. 79), in place of "producing the full argument' " for a law in the existence of which he has by "observation and reflection" been led to believe. For it is not easy to gather from his sketch how he is able to apply a universal law to so varied a series of events and phenomena such as he mentions, and at the same time to point out "the analogy which obtains between the suggested mode of Metazoan development, and the accepted fact of an alternation of generations in the life histories of all plants above the lowest Thallophytes." For in the higher plants the alternation of generations referred to occurs with constancy as regards period of life history, and varies only slightly within the limits of the same group.

Dr. Beard alludes, I presume, to one form of alternation of generations-that of sexual with asexual generation only, or Metagenesis. This he asserts constitutes a general law in the development of Metazoa.

In a sense this may be true enough. If, for instance, we regard the division of each cell as a new asexual generation, then Metagenesis is a very constant phenomenon amongst Metazoa. In this case the life history of a Metazoon consists of a sequence of thousands of asexually produced generations alternating with one sexually produced generation, which gives apparently a stimulus for another run of asexual generations in which polymorphism and division of labour are exhibited in extraordinary complexity and beautiful harmony.

as

But this is not at all what Dr. Beard means. The series of instances which Dr. Beard gives, or system of "nursing Steenstrup termed it, is at most a series of disconnected phenomena of frequent occurrence, and not a law.

Because most Metazoa possess eyes, it is not therefore a law of Metazoan development that eyes should be developed. Diversity in form, number, and time of appearance of eyes, is sufficient to show that the law cannot exist; so also is it in the cases of nursing to which Dr. Beard alludes, and on which he bases his argument.

It seems to me that no "law of alternation of generations in Metazoa can be "enunciated " unless there is evidence forth.

coming of its constant action at corresponding periods in life histories of all animals of different groups, and in a cha similar manner in individuals of one and the same group, a law of such a nature, if it is to be found to act unive amongst Metazoa, must surely have come into action at a early period in the evolution of Metazoa.

Metagenesis is of constant if not universal occurrence cycle of life of Protozoa. A long series of generations. duced asexually is followed by a generation produced ser} that is, a generation produced by the conjugation of tw viduals; this is followed again by another long sere asexually produced generations, and so on. If this is so b

among unicellular organisms of the present day, it is ac unreasonable to suppose it was common among the pr ancestors of the Metazoa and of the Plants. If we are to any form of Metagenesis as a universal phenomenes Metazoa, it must be to the most protozoon-like sage development of the Metazoon that we should look.

There is but one strict meaning to the phrase generation, and that is a fusion of two cells. If Metage means anything it means the alternation of a generation rest. from the fusion of two cells, with one or more generatio sulting from the division of cells.

This we can perhaps find in the protozoon-like ste Metazoan development, and in a way analogous to the alter of generations among plants.

Spermatozoon and ovum fuse and form the fertilized a which is the true sexually produced generation. This p by division a vast number of cells, and if we regard these a number of generations then Metagenesis is obvious ex But it is no more metazoic-if I may use such a wordthe whole animal resulting from the segmentation fertilized ovum, the sexually produced generation.

This is

This generation buds off the immature ovum. the "Primitive ovum" of the embryo. I see no reas this may not be regarded as a distinct asexually poo generation-like the formation of the spore of the plant.

The immature ovum divides into two cells-first Polach and more mature ovum. The more mature ovum divide two cells, namely, second Polar body and mature ovin does not materially affect the argument whether we sho gard these two processes as two separate consecutive asc produced generations, or as one asexually produced t cellular generation. If we take the latter view, the maturation of the ovum is more analogous to the prob stage of the life history of plants.

In either case the result is the formation of the mature comparable to the oosphere of plants.

The mature ovum fuses with the mature spermatococ the sexually produced generation recurs, and the cy development is completed.

I cannot help thinking that if Dr. Beard wishes to disc law of Alternation of Generations applicable to the whole Metazoa, he will find a more favourable hunting ground a those stages of development at which the several gr Metazoa approximate, than amongst those stages wher are farthest apart; and also Dr. Beard will find the a between the supposed Metazoan law and the accepted the vegetable kingdom closer than he could ever hope k it if he continues his present line of search.

If the above theory of the cycle of Metazoan life can be sidered tenable, we see that both in the Higher Plants an Metazoa there are constantly alternating "sporophyte gamophyte " generations, and further, we can find ev as we should expect to do, of the origin of such a t phenomenon in the single celled or protozoan life, whe continuance of the species may be secured in both the namely, by the formation of asexually produced spores, a consequence of cell fusion, i.e. conjugation.

RIC. ASSHET“

Oxygen for Limelight. THE employment of oxygen for limelight and other p¦ has increased enormously since the commercial introduc the Brin method, by which the gas is separated from stuc air by a now well-known chemical process. The gas so is practically pure, analysis showing that as now supplied" Brin companies it contains on an average 95 per ce oxygen, the remaining five per cent. consisting of inert ti

The success of this comparatively new industry has been so ked, that, as a natural result, competitors with rival proes have come forward. Some of these met with failure at early stage of their career, but others are supplying oxygen he public. This is by no means a state of things to be lored from the consumer's point of view, if the product from one source is as good as the other, for benefit generally arises a healthy competition. But when the rival product turns out e not oxygen, but a half and half mixture of oxygen and with a slight excess of the latter, the competition is of a idedly unhealthy character, and is correspondingly bad for consumer. I recently obtained a sample of gas from a er, which on testing (with a Hempel absorption pipette, ged with metallic copper and ammonia) I found to be a ture containing only 60 6 of oxygen. I next tested the ninating value of this highly-diluted oxygen with a limelight and for sake of comparison, placed by its side a precisely ilar jet supplied with Brin's oxygen, and, as might have been ected, the light given by the former was little more than one'as intense as that afforded by the latter. With the good

gen the lime cylinder was quickly pitted, whilst the other wed no symptom of destruction. It is also to be remarked ́t the consumption of the diluted gas was, for a given period, ut one-third more-striving with both jets to get the best sible light-than that of good oxygen. On the same principle ountaineer at a high altitude will pass more (rarefied) air ough his lungs than he will when he is in the valley breathing t which contains the normal quantity of oxygen.

As this matter is of great importance to many workers, I it that you may be able to find room in your valued publicafor these words of necessary caution.

T. C. HEPWORTH.

5, St. Augustine's Road, Camden Square, N. W., December 6.

THE STAR OF BETHLEHEM. the Astronomical Journal of November 26 we find the second of two very interesting articles written Mr. J. H. Stockwell, bearing on the chronology of tain ancient events. In the introduction the author cusses and sums up some of the more important and torical dates which he has determined by calculations ancient eclipses. He next refers to the help which may obtained in the same direction by means of calculations conjunctions of the planets, and quite appropriately the present season points out that the appearance of star of Bethlehem may have been due to the conjuncn of the planets Venus and Jupiter, instead of Saturn 1 Jupiter, as was suggested on incomplete data by pler nearly three hundred years ago. We cannot do ter than lay this part of Mr. Stockwell's communica1 before our readers.

Although the heliocentric conjunctions of the planets ur with a considerable degree of regularity, and are › very easily calculated, the geocentric conjunctions subject to many inequalities in the periods of their cessive occurrences; so that it requires somewhat orate computations to determine accurately the racter of any geocentric conjunction of two planets ch occurred in ancient times. On account of the uency of planetary conjunctions, and the indefinite ner in which they are usually described, it becomes atter of very great difficulty to identify any particular junction unless it is associated with some other event se data can be independently determined. A rekable case of this character is given in the Bible, for thew informs us in the days of Herod the King re came wise men from the East to Jerusalem say "Where is he that is born King of the Jews? for we seen his star in the East, and are come to worship "From the subsequent inquiries and mandates of od the King concerning the time when the star ared, we are led to infer that its appearance took e within two years preceding the death of Herod,

and it has been sought to explain the appearance of the star by means of a conjunction of the planets-the Creator employing celestial phenomena to proclaim the good tidings of great joy, which shall be to all people.'

"The illustrious Kepler was the first to suggest that the star of the wise men might be explained by means of a conjunction of the planets Jupiter and Saturn, and he even undertook to calculate the times when such conjunctions took place. Much has been said and written on the subject of the 'star of the wise men' during the past few years; but no important contribution to the natural history of the star has been made since the days of Kepler, nearly three hundred years ago. But the supernatural history and functions of such a star have been discussed in a very able and interesting manner by many writers in theological, literary, and semi-scientific periodicals during the past twenty years, and perhaps nothing of interest and importance can now be added to what has already been published on that subject.

"I find, however, that Kepler overlooked one important element of the problem in his calculations, and consequently left the natural history of the problem in an incomplete and unsatisfactory condition. I shall therefore here attempt to complete more fully what Kepler began, and show that the Biblical narrative concerning the 'star in the east' is better satisfied by a conjunction of Venus and Jupiter than by any of the conjunctions computed by Kepler.

"We have already seen that the death of Herod took place early in the year B.C. 4, and if we can now show that there was a very conspicuous conjunction of two bright planets, visible only in the east, within two years preceding that date, the hypothesis that such conjunction was the event referred to in the Biblical narrative will at least be rendered plausible, if not entirely legitimate; and for this purpose I have here undertaken the calculation of all the conjunctions of the planets which took place near that epoch. I shall first enquire whether there was a conjunction of the planets Jupiter and Saturn about that period of time which would satisfy the required conditions. The mean interval between two heliocentric conjunctions of Jupiter and Saturn is 7253 4638 days; and they were in mean conjunction B.C. 6, January 30. Now the time of true heliocentric conjunction may differ from the time of mean heliocentric conjunction by 241 days, on account of the inequalities in their elliptic motions, and by 23 days more by reason of the great inequalities of long period in their mean motions. But the time of geocentric conjunction of Jupiter and Saturn may differ from the time of heliocentric conjunction by 102 days; so that a geocentric conjunction may occur one whole year before or after the time of mean heliocentric conjunction. In the present instance I find that the true heliocentric conjunction took place B.C. 7, September 23, which is 129 days before the mean heliocentric conjunction; and that there were three geocentric conjunctions during the year B.C. 7, which took place as follows:

"The first conjunction took place June 7, in which Saturn passed 1° 4' to the south of Jupiter; the second conjunction took place September 18, in which Saturn passed 1° 2' to the south of Jupiter; and the third conjunction occurred on December 15, in which Saturn passed 1° 8' to the south of Jupiter.

"In the first conjunction the pianets would have an elongation of about 73° to the westward of the sun, and would be seen during four or five hours in the east in the morning. The second conjunction took place near the time of opposition with the sun, and would be visible during the whole night, so that it could not properly be designated as a star in the east any more than a star in the west. In the third conjunction the planets would have an elongation of about 84 to the eastward of the sun, and could therefore appear only as evening stars

in the west. Moreover, Saturn is not an especially bright planet, and consequently no one of these three conjunctions could have been very conspicuous in the heavens. The first conjunction was the only one that was visible in the east, but it occurred nearly three years before the death of Herod; it could hardly be said to satisfy the conditions required by the narrative. No other conjunctions of Jupiter and Saturn could possibly occur till about twenty years later, so that we may conclude with a light degree of probability that the phenomenon alluded to in the Bible was not occasioned by a conjunction of Jupiter and Saturn. Since the planet Mars is a conspicuous object when near its opposition with the sun, it may be well to inquire whether a conjunction of Mars and Jupiter might not occasion the phenomenon referred to. But since Mars is conspicuous only near its opposition with the sun, it is evident that any conjunction when in that direction would appear as a star in the west as much as in the east, and consequently it would not fulfil the required conditions. There was, however, a conjunction of Mars and Jupiter on March 5, B.C. 6; but at that the planet's elongation was only 18° to the eastward of the sun, and consequently could have been visible only in the west. But Mars was then so far from the earth, and so nearly in conjunction with the sun, that the conjunction would be wholly invisible. At the same time Saturn was not very far from Jupiter, and hence it was said there was a triple conjunction of the planets Mars, Jupiter, and Saturn in the spring of B.C. 6. "It is evident without calculation there could be no conspicuous conjunction of Venus and Mars at any time; because Mars is not a conspicuous planet unless its elongation from the sun be greater than the greatest elongation ever attained by Venus, so that it would be a waste of time and labour to enter into the computations of any such conjunctions.

"It now remains to inquire whether the two brightest planets of the solar system, Venus and Jupiter, might not have been in conjunction within a short time before the death of Herod, and constitute the phenomenon alluded to in the biblical narrative; for it was the beautiful phenomenon presented by these two planets when in conjunction last February that suggested this investigation. Now the conjunctions of Venus with the sun occur with great regularity at intervals of about 584 days, while those of Jupiter at intervals of 399 days. Moreover, it may easily be shown that all geocentric conjunctions of Venus and Jupiter must take place within about 60 days before or after Jupiter's conjunction with the sun. Therefore, by tabulating the times of Jupiter's conjunction with the sun, we have only to investigate the longitude of Venus for a period of 60 days before or after that event in order to determine whether a conjunction of those planets will then take place. Now I find Jupiter was in near conjunction with the sun B.C. 6, March 29, while Venus was in conjunction on the preceding November 5, or 144 days earlier than Jupiter. Venus was therefore past her greatest western elongation, and was moving towards her superior conjunction, and she would overtake Jupiter on May 8, when their mutual elongations from the sun would be 27 44' to the west. At that time the heliocentric latitude of Venus and Jupiter were 3° 21′ and 1° 20' south, while their geocentric latitudes were 1° 40' and 1 8' south respectively. It therefore follows that at the time of their geocentric conjunction Jupiter was only 32', or about the angular breadth of the moon to the northward of Venus; and as they were then to the westward of the sun, they would be visible only as a star in the east a couple of hours before sunrise. These two brightest planets in the sky would therefore at the time of conjunction, B.C. 6, May 8, be apparently very close together and produce a striking and beautiful appearance. The date also at which it took place being about 50 days less than two years before the death of Herod, harmonizes well with

the spirit and other conditions of the narrative; for is probable that the mandate for the slaughter of the children of two years old and under was issued se months before his decease, and the limit of two yer would leave an ample margin for any uncertainty as the time of the appearance of the star as related by: Magi. "There were no other conjunctions of Venus and Jupeuntil the year B.C. 2, or nearly two years after the de of Herod, when there were two conjunctions, one of wha occurred on August 31 and the other on October 4. Tæ first of these was invisible on account of being too ter the sun; but the second took place when Venus . nearly at her largest elongation to the westward of tr

sun.

"If the preceding calculations, and the references base on them, are correct, it follows that Christ was boma early as May in the year B C. 6; and if He was crucis. at the time of the paschal full moon, which occurred a Friday, it must have taken place on April 3, in e year A.D. 33. And since any given phase of the most s repeated on the same day of the week, and also with two days of the same time of the year, at intervals d 334 lunations, or 27 years, it follows there was no pascha full moon on a Friday between the years A. D. 6 and A 60, except the one on April 3, A.D. 33; whence it wo seem to follow that Christ was thirty-eight years old :: the time of His crucifixion and death, and this w02vindicate the sagacity of the Jewish doctors, who a recently affirmed that He (Jesus) was not then fifty forty years old."

FUJISAN.

ALL who remember the beautiful plates illustrating the volume on "The Great Earthquake of Japan, 1531, which was issued by the same authors a few months 2. will welcome the first instalment of a work wh! promises to illustrate, in a manner worthy of the subject, the magnificent volcanic phenomena of Japar The present part contains ten plates, and is devots: the illustration of the most famous and beautiful of a! the Japanese volcanoes-Fujisan. The number of partthat the authors will publish will depend partly, we ar told, on the encouragement they receive, and partly or the number of photographs that they have been able secure during the past summer.

The photographs in the present part, which are all reproduced as permanent collotypes, 11 inches by 8 inchein size, are exquisite examples of what can be accom plished by this method of illustration, and show that lep.. is certainly not behind any country in the world so fa' as the resources of the publisher go. Where all are so excellent, it is difficult to select any particular plate for especial praise, but one of the most remarkable is cer tainly Plate II., which gives a view over the great cloud banks as seen from the summit of Fuji. Nothing can more striking than the manner in which the effect of the great fleecy masses of vapour are reproduced, and here nothing whatever is lost from want of colour. The plate of greatest scientific interest is perhaps the last, which shows the interior of the crater of Fuji- a great pit neu to 700 feet deep, with perpendicular walls. The sides are built up of rings of variously-coloured rocks, while snow rests in the sheltered hollows. The remaining pictureillustrate the sacred mountain as seen from different points of view, the graceful curves of its outline, the variation in the distribution of snow on its flanks, and the

1 "The Volcanoes of Japan. Part I. Fujisan." By John Milne, F.R.S Professor of Ming and Geology. Imperial University of Tapan, and W K. Burton, C.E., Professor of Sanitary Engineering Imperial Univ (TSC) of Japan. Plates by K. Ogawa (Yokohama, Shanghai, Hongkong, at Singapore: Kelly and Walsh, Limited, 1892.)