obtained with end-on tubes, from 5 to 10 mm. in diameter and about 30 cm. long, with quartz windows, and at pressures in the neighbourhood of 5 mm. The necessary exposure varies from five minutes to an hour, according to the width of the slit, the absorptive power of the medium, etc. I have obtained a beautiful photograph of the absorption spectrum of benzol vapour with fifteen minutes' exposure. E. P. LEWIS. University of California (Department of Physics), Berkeley, California, May 18.

The Relation between Chromosomes and Sex-
determination in "Abraxas grossulariata."

IN a paper on this subject in the Journal of Genetics (vol. iv., June, 1914, p. 1) I gave evidence that in a strain of A. grossulariata which I have bred for several years two kinds of eggs are produced, having respectively twenty-eight and twenty-seven chromosomes. Since the somatic chromosome-number is fifty-six in the male and fifty-five in the female, it seemed evident that the eggs with twenty-eight were male-determining, those with twenty-seven female-determining. In this strain some families in each generation consist entirely of females, so it was hoped to prove the correctness of the conclusion with regard to sex-determination by finding that in families consisting entirely of females all the eggs contain twenty-seven chromosomes. I have now examined the eggs of several such families, and find, contrary to expectation, that the equatorial plate of the inner polar spindle contains twenty-eight chromosomes about as frequently as twenty-seven. The new material confirms the observation that twenty-seven occur in one spindle and twentyeight in the other, but it seems to make it certain that the presence of twenty-eight chromosomes in the inner spindle does not necessarily cause the production of a male-at least, in the strain which produces all-female families. A possible explanation of the anomaly is that in all-female families a chromosome is eliminated at a later stage, but at present I have no direct evidence for this. I have material preserved in the hope of testing this suggestion, but the investigation is likely in any case to be a lengthy one, and circumstances may prevent my continuing it for some months. I therefore make this short statement of the facts as at present known, in order that it may not be assumed that the existence of male- and female-determining chromosomes has been finally demonstrated in Abraxas.

LEONARD DONCASTER.

Zoological Laboratory, Cambridge, June 7.

Cavities due to Pyrites in Magnesium Limestone. IN some districts it seems that iron pyrites formed an important constituent during the deposition of the Magnesium Limestone-not only in the north of England, but in America also in some of the Magnesium Limestones of the Cambrian age.

The Durham beds at Fulwell Hill Quarry give us ample evidence of this, as regards the English beds, by their very numerous cavities, the shape of which, which cannot be attributed to anhydrite, affords the clue. Unless carefully sought, the salts of iron are not noticed, nor are ferruginous band-stains conspicuous or frequent.

Some years ago I obtained specimens from the Roker Cannon-ball bed; some of them, recently placed in the Jermyn Street Museum, have been examined by Mr. A. F. Hallimond, the assistant-curator, who

has labelled them "pseudomorphous after mispickel." These, however, were formed on the outside of the calcareous spheres. I have also secured from the Fulwell Hill Quarry a few specimens of the concretionary structure with cavities containing limonite in powder. There are, however, a large number of empty cavities that apparently once contained pyrites, which are free from the prevalent "marl" powder, from dolomite, or from any traces of anhydrite. They are from about in. up to about 6 in. in diameter, are roughly spherical, but with projecting cones; they are often decorated with strings of white calcite, though occasionally they are ironstained (see Fig. 1).

Rarely, more or less solid speciare met

mens

with (Fig. 2) which

are casts

FIG. 1.-Bryoidal mass (section) with cavity. x1.

of similar cavities, due to the deposit in them of calcium carbonate. Dr. J. Lister, of the Technical Institute, Tunbridge Wells, has kindly examined these for me, and has found them to contain a notable amount of calcium fluoride, which, so far as I aware, is a new observation. Other kinds of cavities are occasionally met with in the concretions of these beds, but, except the socalled cells, there are no others of frequent occurrence, while these, to which I am directing attention, are found in all the beds, both of limestone and marl. I have seen no traces of copper.

am

FIG. 2.-Calcite cast of a cavity probably after FeS2. XI.

GEORGE ABBOTT.

2 Rusthall Park, Tunbridge Wells, May 12.

POISONOUS GASES IN WARFARE AND THEIR ANTIDOTES.

IN

N the Concise Oxford Dictionary a "Stinkball" is defined as "a vessel containing explosives, etc., generating noxious vapours, used formerly in naval warfare and still by Eastern pirates." The Germans have shown the world how science may be degraded in its application commanders have now to deal with a new weapon to the purposes of the pirate, and our military previously unheard of in the field. Steps have already been taken to provide protection for our men, but a survey of the whole question of the composition and the properties of the gases which

have been or are likely to be used may be of service at the present time.

The chemical laboratory would be able to provide a very large number of offensive and poisonous gases and vapours, but for practical purposes they would be limited to those which present the following qualities: (1) they must be much heavier than atmospheric air; (2) they must be producible in large quantity in the form of a portable liquid or solid, which in its turn will evaporate rapidly from the containing vessel or otherwise so as to produce the gas; (3) they should not be excessively soluble in water, or much would be lost in rolling over moist ground.

In the newspaper accounts from eye-witnesses of the cloud of gas sent out by the enemy it has several times been described as presenting a reddish colour. If this statement is based on correct observation it is certain that the gas used on these occasions must have contained either bromine or peroxide of nitrogen, both of which have an orange-brown colour. It seems improbable that oxides of nitrogen would be used on account of the cost in the form of nitric acid or nitrate from which they must be produced. Bromine, however, is made from the salts in the Stassfurth deposits, and from this source up to the outbreak of war much of the bromine and bromides of commerce was obtained. Bromine is at common temperatures a liquid, but it evaporates very easily and produces a vapour which is about five-and-a-half times heavier than air. Probably the amount of bromine available would be insufficient for the production of the enormous quantities of vapour required in these operations if used by itself, but as the whole object of manufacture in the present case is not the production of pure bromine but of something that will suffocate, the material used may consist essentially of the closely allied element chlorine, accompanied with a quantity of bromine sufficient to account for the colour.

At ordinary temperatures chlorine is a pale-green gas which is about two-and-a-half times heavier than air. The gas was discovered by the Swedish chemist Scheele in 1774, and has been used for bleaching purposes since the end of the eighteenth century. It acts rapidly on nearly all metals if moist, but when free from water it does not attack the surface of iron, and as it is easily reduced by compression to a liquid, it has been produced commercially in very large quantities for many years past in the liquid state and preserved in iron bottles. There can be littie doubt that the gas from which our men have been suffering is sent into the trenches in such cylinders. The gaseous chlorine which escapes from them on simply opening the tap, whether or not it is accompanied with bromine, is quite sufficient to account for the suffering and death which have been the result of getting the gas into the lungs, but other volatile substances have been suggested as possibly available. Thus phosgene, a compound of carbonic oxide and chlorine, is a heavy gas about three-and-a-half times heavier than air, easily

liquefiable and easily vaporisable from the liquid, which is a commercial article produced in Germany on a fairly large scale. Sulphur dioxide, often erroneously called surphurous acid, is familiar as the product of burning sulphur, and being liquefiable readily it may be seen in the liquid form in glass siphon bottles in every chemical laboratory.

The chlorides of sulphur, phosphorus, and arsenic are also very irritating and poisonous. The only reason for supposing that these compounds might be employed is provided by a communication from Warsaw which appeared in the Times of June 5, in which it is stated that the Germans. had been burning straw on which a white powder resembling salt had been sprinkled. The dense smoke carried by the wind over the Russian lines is stated to have produced symptoms similar to those reported from France in the case of victims of suffocation from what is believed to be chlorine or a mixture of chlorine and bromine. The white powder used against the Russians probably contains one of the chlorides mentioned above, which might be made portable by admixture of common salt. The statement in the Times that the powder is believed to be some easily-made compound of chloral is obviously a misprint.

Fortunately all these fumigating agents agree in one particular. They can be absorbed and therefore stopped by passage through or over a strongly alkaline substance to which may be added, especially in reference to chlorine or bromine, a proportion of sodium hyposulphite (thiosulphate), the familiar "hypo" of the photographer. It is important to notice that a strong alkali is necessary and in layers sufficiently thick. When the ordinary housekeeper speaks of "carbonate of soda," the salt known to the chemist as bicarbonate is always intended, and this is almost useless.

The masks or respirators supplied to the troops consist of material saturated, though not dripping, with a strong solution of common washing soda (the carbonate) mixed with an equal quantity of hyposulphite, to which has been added 2 or 3 per cent. of glycerine to keep the whole damp. A very good material would also be the granular mixture of lime and caustic soda, known in every chemical laboratory as "soda-lime." This would have to be wrapped up in gauze and would not require to contain glycerine.

Many well-intentioned efforts have been made by private persons to supply their soldier friends with respirators made of gauze or muslin containing a receptacle filled with cotton waste saturated with an alkaline solution. Most of these attempts are not only imperfect but really dangerous to the men by leading them to consider themselves protected when practically there is no protection. The pads saturated with alkali are often much too small, and the cotton padding has not been secured in position by proper quilting or otherwise. Moreover, the most efficient solution has not been used. The official protectors take the form of a hood which covers the head and afford a complete protection against the gases, even

when concentrated. As they are now being sent to France daily in very large numbers it is to be hoped that the directions of the War Office recently issued will be attended to and no more respirators of other types, either home made or manufactured without official sanction, will be sent to the Front.

It must not be forgotten, however, that the efficiency of all these contrivances is limited, and in the event of being exposed to several successive doses of gas the material ought to be washed in clean water and then recharged with the original chemical mixture. A supply of liquid is now furnished to the men in the trenches.

A question has arisen in connection with the idea of the possible use by the enemy of bombs containing poison, to be dropped by Zeppelins or other aircraft. In the event of such occurrence the best course would be in the first instance to

Ignition Device

mittee. As the committee points out, fires are not usually caused by the ordinary explosive bomb. When fire occurs it is due to open lights or fires in the house or to broken gas or electric mains. The incendiary bomb is, however, designed to produce a fierce fire in itself. The accompanying diagram shows the construction of one of the ordinary types of these missiles.

The fumes from such bombs contain a large quantity of phosphoric acid, and when inhaled produce violent coughing, but fortunately they would be completely stopped by a wet cloth. Thermit, in its ignition, produces such a high temperature that the timbers of a house would be certainly inflamed if struck, but when the first burst of flame is over the extension of the fire would proceed in the manner of fires originating otherwise and would have to be dealt with by the fire brigade in the usual way. In the meantime, every house in a threatened area should be furnished with buckets and baths full of water ready drawn to be applied as quickly as possible, in quantity as large as possible, in case of necessity. W. A. TILDEN.

Thermit

SECTION OF INCENDIARY BOMB.

Retinous matter

Melted white phos phorus

(a) The bomb, as a rule, is conical, of 10 in. diameter at the base, corded round, and has a metal handle at the apex (see A). (6) The base is a flat cup, on to which a pierced metal funnel is fitted, having the ignition device and handle fitted at the top. (c) The funnel is generally filled with Thermit, which upon ignition generates intense heat, and by the time of the concussion has taken the form of molten metal of the extraordinary high temperature of over 5000° F. The molten metal is spread by the concussion. (d) Outside the funnel is a padding of a highly inflammable or resinous material bound on with an inflammable form of rope. The resinous material creates a pungent smoke. (e) There is generally some melted white phosphorus in the bottom of the cap, which develops nauseous fumes. (f) In some cases celluloid chippings are added and occasionally a small quantity of petrol.

close all the windows and doors of houses in the neighbourhood. Should the smell of chlorine be perceived indoors, a cloth wetted with strong solution of ammonia should be waved about in the air. This would produce a harmless white smoke consisting of sal-ammoniac, and this, even 'with excess of ammonia, would produce very little if any damage to furniture, etc. With regard to wounds produced by shrapnel or other projectiles containing phosphorus, it is improbable that any of the element would escape combustion in the air. The mischief is caused by the phosphoric acid produced being carried into the tissues, and for this dressings made slightly alkaline with carbonate of soda should be used till all acid has been removed.

The chief result of air raids so far has been the production of fires, and a timely warning has been issued by the British Fire Prevention Com

VISIBILITY.

THE factors which determine the ability of the eye to distinguish an object from its background are not very well known for many of the conditions met with in practice. Our practical experience has taught us that if we wish to see properly under any given condition, the illumination must not be less than a certain amount, and we have been content to provide the necessary illumination without having any very clear idea what are the aids and hindrances to good seeing. A discussion on some aspects of the subject of visibility was initiated at the Illuminating Engineering Society, on April 27, by Messrs. Paterand Dudding, of the National Physical Laboratory; and many speakers made interesting contributions to it.

son

Several fields of investigation disclose themselves in connection with the subject, and are mainly concerned with cases in which objects become difficult of discernment. For instance, the conditions required to render troops or other objects invisible to the enemy is a matter which at first sight may appear simple, but if all that is said is true, the problem is a complicated one, and would justify thorough research into it.

It has been stated that if a donkey and a zebra are located in a distant field (and remained there to be observed) the zebra would become invisible long before the donkey. This effect was discussed in an interesting article by Col. F. N. Maude in Land and Water for January 30, in which he gave an incident from practical experience of a regiment with pipeclayed belts and accoutrements being invisible by the side of another regiment which had equipment of a more uniform nature.

The question of the visibility of distant faint. sources of light at night is the simplest form in which the problem presents itself, because first, the contrast between the light and its background

A

is very great, and secondly, because the object viewed is below the resolving power of the eye (about one minute of arc), and therefore the dimensions of the object_viewed have not to be considered. The visibility of such point sources of light is inversely proportional to the square of the distance of the observer from them, and directly proportional to their candle-powers. Their visibility becomes an important matter at sea, since the navigation of ships so as to avoid collision depends almost entirely on them. practical working visibility is 16 candles at two sea miles, whether the light be red, green, or yellow. There is a very marked difference, however, in the relative behaviour of these lights to oblique vision. To normal people a green light will appear some five times brighter when viewed obliquely than when viewed directly, but a red light acts in exactly the opposite way. If we place a point source of green light in a dark room so that its visibility from the position viewed is equivalent to about o'or micro-candle at a metre, most observers will pick up this light when looking in some other direction and be quite conscious of its presence, but no sooner does one look towards it than it is gone. A faint red light, on the other hand, cannot be picked up by oblique vision at all, but when an observer has finally discovered it by direct vision he is impressed by how very visible it is. If he glances to one side, however, it disappears. The colour of faint lights is not distinguishable by oblique vision.

Under favourable laboratory conditions of threshold visibility it is calculated that the energy which is just sufficient to stimulate a single element of the retina of the dark adapted eye is at the rate of about 10-15 watt.

The question of the visibility at night of objects of finite size is not so simple. The problem occurs in a practical form in the showing-up of an obstacle by a motor headlight, such, for instance, as a pedestrian on the road ahead of a car. Such objects subtend angles well above that of the resolving power of the eye, and there is no justification for assuming that their visibility follows the same simple laws which hold for point sources of light. The experiments show in effect that the visibility of objects subtending angles up

to about ten minutes of arc follows the same law as for point sources, viz., the brightness must be inversely proportional to the square of the distance, but that for objects above this the brightness is proportional simply to the distance. It is the larger size of objects which have generally to be discerned when driving at night, and it therefore follows that for equal visibility at twice the distance the equivalent candle-power of a headlight must be increased eight times, and not sixteen times, as would be the case if the inverse square law held.

Thus it follows that the useful range of a motor headlight is proportional to the cube root of the intensity of the beam.

The visibility of objects viewed in the presence of glaring lights is a matter which has been much discussed, but the reason for the discomfort and

annoyance which is felt when one has a light in one's eyes is rather obscure-except in such gross cases as the dazzle produced on the dark adapted eye by motor headlights. Experiments show that at ordinary illuminations it is very difficult to detect by measurement any diminution of ability to see detail in the presence of bright lateral lights, although there is little doubt about the discomfort caused by them. When, however, the surface brightness of the objects to be distinguished is very low there is, when lateral lights are present near to the object, a marked falling off in the ability of the eye to distinguish the slight contrasts which exist over such surfaces, and detail appears in consequence obliterated. A room viewed under these conditions might have to the eye somewhat the appearance of an under-exposed negative in which only the high lights and strong contrasts show up. It cannot be said, however, with certainty that herein lies the evil of glaring lights, and the discomfort experienced with them may very well be psychological in character.

The discrimination of detail in certain circumstances depends almost entirely either on shadow or on the direction of the incident light. Where the surface of an object to be viewed is uneven, but uniform in colour, the only way in which the unevenness can be shown up is by differences in the illumination of such surfaces caused by the different angles which they present to the incident light. If the unevennesses are very deep they will be shown up by the shadows which are thrown by the raised portions on to the surrounding surfaces. For instance, the embossed lettering used on some notepaper depends entirely on this action, and a sculpture in bas-relief must obviously present a very different appearance according as it is illuminated by unidirectional light at glancing incidence or by light from a large source striking it mainly at normal incidence.

The question has a wide practical application in the manipulation of self-toned fabrics, viz., unicoloured fabrics with no natural contrasts. A person doing needlework with such materials depends for the discrimination of the detailed strands of the fabric on the small shadows cast by one strand on to the next and on the varying brightness over the curved surfaces of each individual strand. Both these factors depend on the unidirectional character of the light incident on the material, and the inefficiency of thoroughly diffused, viz., indirect light, for such work is most marked. The writer has, for ten years, used indirect lighting for domestic use, and for a long time ridiculed the assertion which was often made that it was a most unsatisfactory light by which to do needlework, and particularly darning. After experiment, however, it is clear that the contention is sound, and that there is nothing so good as a unidirectional light giving harsh shadows for the discrimination of detail in needlework of all kinds. Indirect lighting, with its soft shadows, is an ideal light for domestic use, except in respect of this one particular.

C. C. PATERSON.

OLFACTORY STRUCTURES IN INSECTS.

THERE is no doubt that many insects have a

sense of smell, but there is great variety of opinion as to the precise location of the sense. Dr. N. E. McIndoo,1 of the Washington Bureau of Entomology, has summed up the discrepant views in forty pages, and also in one word"chaos." Lehmann seems to have been the first to experiment (1799), and he was led to the conclusion that the seat of smell is in the spiracles. Most of the older naturalists reached their conclusion without experimenting, and the sense of smell has been referred to at least a dozen different parts, such as the mouth, the epipharynx, the palps, the caudal styles. Of recent years, as the result of experiment on one hand and histological analysis on the other, there has been a tendency to conclude that the antennæ are the olfactory organs. The antennæ are rich in sensory structures, and their removal is sometimes followed by a negative reaction to an odour which is attractive to the intact insect. Dr. McIndoo thinks that the arguments are very inconclusive.

There are certainly many structures on the antennæ which might be olfactory-pore-plates (Lubbock's pits), pegs (Lubbock's cones), Forel's flasks, pit-pegs, and end-rods; and each of these has been claimed by some investigator as the true and only seat of smell. But in all these structures the nerve-ending is shut in by the chitinous cuticle, through which, therefore, the odour would have to pass. Another difficulty phasised by Dr. McIndoo is in regard to the distribution of the structures above-mentioned: thus the pore-plates cannot be the exclusive olfactory structures, for they are absent in all Lepidoptera; the pegs cannot be the exclusive olfactory structures, for they are absent in many male bees, and

em

SO on. The distribution of the various antennary structures in different types does not correspond with the varied rates of response to odours as shown by these types under experimental conditions. Moreover, spiders can smell, and they have no antennæ. The author concludes that for ants, bees, and wasps, the antennæ can no longer be regarded even as a possible seat of the sense of smell. It is possible, however, that what is true of one order of insects may not hold for another.

What, then, is to be made of the experiment repeatedly performed of removing the antennæ and observing that the usual response to odours did not occur? The author's numerous experiments on Hymenoptera have shown him that if the antennæ of these insects are mutilated even a little, the behaviour becomes abnormal, and the slow reaction to odours may be due to the actual injury, not to the removal of some of the olfactory structures. Amputation of the antennæ

1 "The Olfactory Sense of Insects." By Dr. N. E. McIndoo. Smithsonian Miscellaneous Collections. Vol. lxiii (1914), number 9. Pp. 1-63, 6 figs.

is often fatal, and the insect is so much disorganised that its failure to respond to attractive odours does not prove that the olfactory structures are on the antennæ. Details are given in support of this useful criticism.

Where, then, is the seat of smell? Dr. McIndoo takes us back to the work of Hicks (1857), who discovered vesicles or pores on the bases of the wings and on the legs, and suggested that they were olfactory. The structures have been studied by Janet and others, as well as by Dr. McIndoo. Each is like an inverted flask imbedded in the chitin, but with a minute external pore. A fibre from a sensory cell near the inner end of the flask rises to the pore, and its cytoplasm comes into direct contact with the air and its odorous particles. These pores correspond to the lyriform organs or slits discovered by Bertkau on the legs of spiders, and subsequently studied by Dr. McIndoo. When the pores on insects' wings are covered with glue or vaseline the reaction times are greatly increased, and the rates of response in particular insects correspond with the number of pores. A drone hive bee has 2600 pores, and responds in 2'9 seconds; a worker has 2200 pores, and responds in 3'4 seconds; and a queen has 1800 pores and responds in 49 seconds. The author is to be congratulated on his introduction of some order into the chaos of discrepant opinions concerning the seat of the sense of smell in insects.

SIR A. H. CHURCH, K.C.V.O., F.R.S.

THE announcement in NATURE of June 3

of the death of Sir Arthur Herbert Church, in his eighty-first year, has been received with Sir Arthur great regret among men of science. Church was educated at King's College, London, the Royal College of Chemistry, and Lincoln College, Oxford, where he took a First in the Natural Science School. He afterwards became

Professor of Chemisty in the Royal Agricultural College, Cirencester. This appointment naturally led him to devote special attention to agricultural chemistry, on which he became an authority, and at the same time to direct his attention to the chemistry of plants. He contributed memoirs on vegetable albinism; colein or erythrophyll; and aluminium in vascular cryptogams, etc., and known as Turacin, which contains 7 per cent. also investigated the remarkable animal pigment of metallic copper.

Sir Arthur Church also directed his attention to mineralogical chemistry, being the first to discover Churchite, a native cerium phosphate, and several other new minerals; and he was at one time president of the Mineralogical Society. His researches in other departments of applied chemistry seem, however, to have been influenced by his strong interest in art in every form. Perhaps few chemists know that Sir Arthur Church once exhibited at the Royal Academy, besides