CHEMICAL NEWS July 5, 1912 } Drinking Water and Health. 3 disease. But with a lowered vitality when the host of | precipitation of a district and the volume of water which intruders is too great and strong to battle with, they may might be available for a water supply. be beaten in the warfare, and we succumb. Among waterborne diseases the one we have to fear most is typhoid fever. The excretal discharges of its victims are loaded with its bacilli, and when such waste finds its way into a water supply the disease is disseminated, and an epidemic results. Herein lies the chief and great danger in using a supply polluted with sewage or execretal waste. It must, however, be added that water is not the only vehicle which conveys this disease; the ubiquitous house-fly, as we know, must now bear its share of the blame. But there is another danger in impure water, though of this bacteriology takes no note. I refer to the presence of certain poisonous substances, the products of the decomposition of organic matter-either of animal or vegetable origin. There is good evidence that such polluted water may cause headache, nausea, indigestion, diarrhoea, lassitude, and generally lower the vital tone of the system. It is quite true that such toxic compounds have not been Isolated, but I might answer that such is the case with many ptomaines, organic compounds occasionally occurring in our foods-and especially in those which have been stored. Such foods may be, and frequently are, consumed with fatal results. There is every reason to believe that certain waters, and more particularly stagnant waters in which there is decaying vegetable and animal matter, possess this poisonous property. Some of us may have experienced the nauseating effects of water from a pond or lake containing the products of decaying algae. It is scarcely necessary to add that such water is unfit for consumption. Moving water is, as a rule, free from this class of impurity. This is a phase of the water question that has not received from sanitarians the attention it deserves, but I am convinced of its importance in judging of the merits of a water for a city or house supply. So far we have learnt that what we have to fear in our water supplies is, first, the presence of disease germs, due to contamination with sewage, and secondly, those products of the decay of organic bodies from certain classes of matter, excretal or vegetable, and which exert a toxic action on the system. A third form of pollution met with is the waste waters of manufactories which are run into the water course without proper purification. These refuse waters may contain organic or inorganic substances detrimental to health. Fortunately in Canada this kind of pollution is not often found, but in the protection of our lakes and rivers legislation must take cognisance of it, and the laws preventing the discharge of such waste into possible sources of water supplies rigidly enforced. In considering the role of rain and snow in Nature some two years ago, we learnt two facts of a fundamental character. The first was that the earth's moisture was in continual circulation. The ascension of water in the form of vapour, due to the heat of the sun, went on constantly, day and night, winter and summer, from earth and water surface alike. Ice and snow, as we saw, could be converted into vapour without visually passing through the liquid state. This vapour of water ascends until it reaches the higher and colder strata of the atmosphere where it is condensed to fall as rain, hail, or snow, according to the atmospheric conditions prevailing at the time of the precipitation. This process of evaporation and condensation -distillation, in fact-is from the point of view we are considering to-night one of the greatest importance, for it is primarily one of purification. The sun, then, is the agent above all others that renders it possible to obtain a whole. some supply of drinking water, for the water in being con. verted into vapour leaves behind all those substancesmineral and organic-which it held in solution, and descending gives us one of the purest forms of water found in Nature. And, secondly, it was apparent that all our water supplies-lakes, streams, springs, and wells-were directly dependent upon the fall of rain and snow, and therefore there was a very close relationship between the annual There are two properties of water that must be referred to, if only briefly, in order that we may intelligently consider the various classes of water that are suitable and wholesome for domestic use-its solvent power and its carrying power. Water is known as the universal solvent. It is because of its ability to dissolve gases and solid substances, whether they be inorganic (mineral), or organic, and the constant exercise of this power that in Nature there is no such thing as pure water-that is, chemically speaking. Pure water, as formed in the laboratory, consists solely of oxygen and hydrogen. All natural waters, then, contain dissolved matter, some more, some less, and, speaking broadly, the nature of this matter-whether injurious or harmless to health and its amount, will be determined by the character of the rock or soil it passes over or passes through. Thus we have soft waters from the Laurentian districts because the gneisses and granites are not easily soluble and impart but little mineral matter to the water; and we have hard waters in limestone districts, because the water with the aid of the carbon dioxide it has taken from the atmosphere is capable of exerting a very considerable solvent effect upon such rocks and contains as a result more or less lime in solution. Next to the sun, the soil is Nature's greatest water purifier, for it can remove by oxidation and filtration impurities in solution and suspension, but if the soil is choked with filth then the water in passing through it will dissolve such, and be rendered foul. The carrying power of water is secondary to its solvent power in this consideration of natural waters for drinking and household purposes. The descending rain, the storms, the spring freshets, and floods, wash the surface of the land, and carry much which they find there to the nearest stream or lake. Similarly, the banks and channels of streams are eroded-even rocks may be slowly worn away, and the detritus, the débris, borne in the turbid waters, perhaps hundreds of miles, to be deposited as their velocity is checked. In this way deltas of clay and silt, and fine sand mixed with organic particles are formed at the mouth of great rivers, and areas of vast size and of extreme fertility built up. Since turbid waters, those with clay and silt in suspension, are not desirable for supplies, they must be subjected to filtration. If such waters possess no organic filth, the filtered and now clean water will be quite satisfactory. Waters as used by towns or for isolated households, as on the farm, may be classified as follows:-Rain-water, upland surface waters, ground waters or those of shallow wells, and deep-seated waters, as obtained by drilling or boring and among which many springs may be placed. In Rain-water. This can be caught and used as such. As a drinking supply little need be said of this source. Canada, where in most districts other and larger sources of supply are readily available, rain-water is seldom used, save for washing and laundering purposes, for which by reason of its extreme softness it is eminently suitable. Its quality or purity will depend on the condition of the atmosphere through which it falls; if in town we may expect it to contain soot and gases from which it would be comparatively free if falling in rural parts. Again, dirty roofs and eave troughs, storage tanks in which organic débris accumulate, all contribute towards making this supply foul and unfit for consumption—so that even a fairly pure rain-water that has been stored is difficult to find. However, if fresh and clean, it is not at all unwholesome, though not very palatable. If stored in vats or tanks these should be of cement, and frequently examined and cleaned. The water for use should be passed through an efficient filter, and boiling would be an additional safeguard, though the presence of disease germs would not naturally be expected. Upland Surface Waters.-These constitute the waters of our lakes and streams, and are formed by the run-off om the lands, though to some extent, of course, these sources are fed by springs. By far the larger number of | former will be distinctly the inferior water, one that must supplies of Canadian cities and towns are drawn from be efficiently filtered and purified before it can be regarded lakes and rivers, and hence the importance of immediate as a wholesome potable supply. and efficient legislation that will protect these natural bodies of water from sewage and other pollution. The fact should be emphasised that these natural waters are, almost without exception, eminently suited without any preliminary treatment for drinking and domestic use. But as our population increases, and especially as cities and towns build up on the margins of lakes and the banks of streams, the necessity of adequate filtration becomes apparent. It will therefore be the part of wisdom from this on, not only to protect these waters from pollution as effectively as possible, but, also for those communities drawing upon them for their supply to establish filtration plants. Experience in other countries has shown that despite the most viligant protective measures such waters may at any time, through accident or otherwise, receive | excretal waste, and become a source of danger, a menace to good health. It is now generally recognised by the highest authorities that filtration is imperative-a sine qua non-if the supply is at all seasons to be relied on as free from injurious bacterial lite. The nature of the country and the composition of the rocks of the catchment area will largely determine the character of these waters. Thus a limestone district gives rise to a hard water, a Laurentian area, with gneiss, granite, and similar rocks, results in a comparatively soft water. Again, the colour of these waters is largely determined by the presence or absence of swamps in the country from which they draw their supply. A coloured water, that is, one brown or yellowish brown, through the presence of dissolved peaty matter, though offending the æsthetic sense (for we all prefer a colourless water), may be perfectly wholesome, and especially so when such is from a large body of quickly flowing water, as, for instance, the Ottawa river. There are very few cases of illness or indisposition on record-if indeed any that can be definitely traced to the consumption of these peaty waters from large actively flowing waters, provided of course such have proven to be free from excretal pollution. These so-called peaty waters, and from sources such as I have described, have shown themselves almost universally to be perfectly satisfactory for city supplies, not only from their extreme softness (which means a considerable saving in soap and labour to the community), but from the hygienic standpoint. These waters keep well, for their dissolved peaty matter does not readily undergo further decay, is, in fact, remarkably stable. It is true that temporary indisposition frequently follows the use of these waters when one has been accustomed to a hard colourless water, but it is equally true that the reverse happens. Any change in the character of the water consumed may bring about a slight derangement, for the system becomes habituated to a certain water, and some persons are very susceptible, for a time, to any difference in its character. Ground Water.-This is the rain and melted snow absorbed and retained by the soil and subsoil. It is the source that supplies the shallow domestic well so commonly used on the farm homestead and in the village. When the surroundings are perfectly satisfactory from the sanitary standpoint, these wells are frequently a source of excellent water, but when, as is usually the case, convenience to the house or farm buildings is alone considered in the location of the well, the water is seldom of first-class quality, and more often must be adjudged as quite unfit for consumption. On the larger number of farms we find these wells, usually between 10 and 25 feet in depth, sunk in the barnyard or under the stable or other outbuildings, or not very far from the privy (a most crude and unsanitary affair as a rule), or near the back door, out of which the household slops may be thrown, and near which the garbage heap with all sorts of refuse may be found. It is quite true that most soils, and more particularly those that are porous and well aërated (gravels and sands), possess filtering and purifying properties in a marked degree, but the soil surrounding wells located as we have described must in time become saturated with organic filth of a most objectionable character, and is then no longer able to purify, but rather serves to more seriously contaminate the water passing through it to the well, which under such conditions may be said to act as a cess pit. Further, we frequently find these wells become the watery grave for rats, mice, frogs, and other small animals, the decomposing bodies of which render the water foul and unfit for use. Imperfect protection of the mouth of the well may allow the entrance of surface wash. Rotten crib work is another source of contamination. Other causes of pollution could be enumerated, but enough has been said to justify the conclusion that the ordinary farm well is at the best a poor supply, and should be abandoned for a safer, purer source. The examination in the laboratories of the experimental farms of hundreds of samples of such well waters have shown that few of these wells furnish a supply that can be considered wholesome, by far the larger number must be condemned as totally unfit for use. Considering the location of most farm wells it is not a matter of surprise that but a very small proportion of them yield water of sufficient purity to be classed as satisfactory. Many of these waters are colourless, bright, sparkling, clear, and cool, but these qualities are no criterion, and it is by no means uncommon to find waters possessing all these commendable properties and at the same time reeking with filth. Of course, if a well-water becomes turbid after a rain there is reason to reject it, for in this turbidity we have a sign that the soil is no longer able to do its work as a filter and purifier. A precaution of very considerable value towards proThe case, how-tecting the well-water from organic filth is to line the well to a depth of, say, 10 or 12 feet to a thickness of, say, 6 inches with concrete or puddled clay. This lining should project some 6 to 12 inches above the mouth of the well. This prevents the direct inflow of wash and of water trom the surface soil in which the larger amount of putrescible organic matter is found, and ensures a certain amount of filtration through clean layers of soil. ever, with coloured waters from low-lying swampy shallow lakes and ponds is very different. Such bodies of water being more or less stagnant, produce an abundance of vegetable growth largely algal, which under favourable weather conditions may rapidly decompose, giving rise to offensive and nauseating products. If, as frequently happens in summer, these decay products accumulate, in other words get ahead of growth that can utilise them, the water becomes foul and unfit for consumption. The result of drinking such water usually shows itself in an attack of diarrhoea or nausea. From these considerations it would be obvious that colour is not in itself a quality or factor that can be used alone in deciding upon the suitability of a supply. Leaving out of consideration sewage pollution, we may have on the one hand a comparatively colourless water, but one in which alge and other low forms of life are present in large numbers, and in which chemical analysis proves the presence of easily decomposable organic matter, and on the other hand a highly coloured peaty water from a large and quickly flowing river, and the Another safeguard is to keep an area of, say, 50 yards radius round the well free from manure and all deposition of filth (it should preferably be in sod), and this plan we would heartily recommend to those who are contemplating sinking a well for household use or for watering stock. If the ground surrounding the well is an undisturbed area and free from all excretal waste, it will perform its function as a natural filter and the water may be very good. Especially is this the case if the soil is sand or gravel, for such will not only remove suspended matter and germ life, but will also foster the destruction by oxidation of the organic matter held in solution. A clay subsoil is far inferior to sand in its purifying effect. 5 palatable and the "flat" taste removed by being allowed to cool in the open air. And now, in conclusion, I must emphasise two points. The first is the insidious character of polluted water. The danger that lurks in water polluted with excretal products is not always apparent. This fact must not be lost sight of. There may be no outbreak of typhoid fever, but it may be generally undermining the health. In far too many cases the well goes unsuspected until the victim is stricken down. The moral is, ascertain the purity of the supply. Deep Seated Waters.-These are waters that have percolated through the soil and permeable rock strata until arrested by an impervious stratum. They may appear on the surface as springs, but are more commonly obtained by deep wells, driven or bored, possibly through several overlying impervious strata to the water bearing rock. If there are no fissures in these overlying strata and there is no opportunity for water to flow downwards between piping and the sides of the boring, a good water will in all probability be obtained. While it cannot be taken for granted that a bored well will necessarily yield a good drinking water, it is the source of supply to be generally recommended for the isolated households. Examination has shown that they are capable of furnishing in the larger number of instances, and when proper precaution has been taken to exclude surface water, a supply of high organic purity and very low bacterial content. In certain districts we find these deep seated waters characterised by an excess of saline matter, rendering them unsuitable for domestic use; but when such is not the case the deep well undoubtedly constitutes a safer and better source of supply than the shallow, ground water well. With a pump actuated by a windmill, small gasoline or hot air engine, tanks can be filled in the farm buildings for the watering of the stock, and in the farm house to supply the bath room and kitchen. Such an arrangement would mean much, not only in the matter of convenience and the saving of labour, but in By T. CROOK, A.R.C.Sc. (Dublin), F.G S., and S. J. JOHNSTONE, the still more important matter of securing a supply that would lead to better thrift in the stock and better health in the family. Before bringing this Address to a close I must answer, though it may be briefly, one or two questions that have been handed me for reply. 1. Is a hard water injurious to health? The human system has a remarkable adaptability, and though certain authorities have considered that a hard water is inducive to the formation of calculi there is very little evidence to support the statement. Cities having even a very hard water supply do not show the prevalence of any disease that can be attributed to the water, and we may conclude that the lime compounds present do not work any injury to health. As already remarked sudden changes from one character of water to another, whether hard to soft or soft to hard, may cause disturbance in the system, but such will only be temporary. The system requires lime to build up its skeleton and for its other tissues, and it may take it from the water as well as from the food; there is nothing to prove that the lime taken in the water is not as readily assimilable as that in the food-stuffs we consume. Consensus of opinion points to a moderately hard spring water, in which all possibility of contamination is out of the question, as probably the best supply, but such is unfortunately very hard to find. 2. Is distilled water wholesome? The only argument that can be urged against its use for drinking is that it does not contain the necessary mineral elements for the building up of the tissues and for the replacement of the daily outgo of these elements. The answer is that in the ordinary normal diet there is such an abundance of the mineral salts that the absence of them in the drinking water need cause no alarm. There is much to be said in favour of distilled water, as it should be free from all forms of organic matter and disease germs. 3. What means can the householder take towards making a suspicious water harmless? Undoubtedly the best plan is to boil the water for from five to fifteen minutes. This is the most efficient safeguard that can be proposed for the individual. Household filters, though removing suspended matter, are seldom to be depended upon to deprive the water of germ life, and at the best require constant attention and cleansing to be kept even fairly efficient. The addition of hypochlorite of lime, now largely used in the purification of city supplies, is not readily applicable in the house, and cannot be regarded as equal to boiling for the destruction of germs. The boiled water may be rendered And the second point is that there is abundance almost everywhere of pure water. There is no better watered country in the world than Canada. We can unhesitatingly affirm that the normal waters of our lakes, streams, and springs, our ground waters and our deep seated sources, are of the purest. It becomes our duty as communities and individuals to preserve and protect them from pollution, and to see to it that the water we drink is as irreproachable in quality as that with which Nature has supplied us. ON STRÜVERITE FROM THE FEDERATED Scientific and Technical Department, Imperial Institute. General Remarks. THE mineral dealt with in this paper was sent for examination to the Imperial Institute by Mr. J. B. Scrivenor, Government Geologist to the Federated Malay States. It occurs on the river Sebantun, about half a mile above Salak North village, Kuala Kangsar district, Perak. The ground on which it was obtained was held on a tin-mining lease, but had to be abandoned owing to the presence of the unknown mineral, which rendered mining unprofitable. Mr. Scrivenor has ascertained that the locality is occupied by a small alluvial flat, that the mineral in question forms the bulk of the concentrate obtained from the alluvium, and that cassiterite, monazite, topaz, tourmaline, zircon, and iron pyrites also occur. The mineral was originally supplied to Mr. Scrivenor by Mr. R. L. Corbett, who stated that he had obtained it by magnetic separation with a Wetherill machine, using a current of from 70 to 75 volts and 10 to 14 ampères. As received at the Imperial Institute the sample consisted of coarse angular grains, 4 or 5 millimetres in diameter. They appeared to consist entirely of a lustrous, black mineral. On looking over the specimen with a lens, however, some grains of a more brownish appearance could be picked out, and these proved to be cassiterite. Occasional particles of quartz were also observed to be intimately associated with the black, lustrous grains. titanium, tantalum, and iron were the essential ingredients. A preliminary analysis of the specimen indicated that This composition, taken in conjunction with the physical characters of the mineral, led to the conclusion that we were dealing with a specimen of strüverite, as defined by Drs. Prior and Zambonini (“On Strüverite and its Relation to Ilmenorutile," Mineralogical Magazine, 1908, vol. xv., pp. 78-89). The analytical difficulties, however, made the satisfactory chemical proof of this a very tedious piece of work. Physical Characters. In the mass the mineral is black, and has a somewhat lustrous appearance. The specific gravity, determined on an amount weighing about 13 grms., which had been freed as far as possible from fragments of cassiterite, was found to be 5:30. The streak is not quite black, but has a somewhat greenish tint. This feature suggested the probability that the mineral was not likely to be quite opaque in microscopic splinters. An optical examination of the fine crushings in this par ticular case gave interesting results. As far as can be judged from the appearance of the small particles obtained by fine crushing the mineral seems to be homogeneous. These particles are not opaque, and they show strong pleochroism. Examined with one nicol most of the thin flakes show a change from brownish-yellow to dull bluishgreen on rotating the nicol. These flakes are birefringent, and compensate with a gypsum-plate when the fast vibration-trace of the plate lies along the maximumabsorption vibration-trace of the flake. Occasionally flakes are observed which show only the brownish-yellow colour on rotating the nicol, and these act isotropically. The optical behaviour thus suggests that the mineral is uniaxial, and on this assumption the pleochroism for the thinnest flakes may be defined thus: o brownish-yellow, e=dull bluish-green. Thicker flakes show a change from brown to black, somewhat resembling that of brown tourmaline, but in the reverse sense, the ordinary ray being less absorbed. In describing the physical characters of strüverite from Piedmont, Zambonini states that the mineral is opaque even in the thinnest flakes. His description of the mineral powder as grey-black," however, may be taken as an indication that the mineral is not quite opaque. In a recently published paper by Hess and Wells ("An Occurrence of Strüverite," Am. Journ. Sci., 1911, Ser. 4, vol. xxxi., pp. 432-442), Hess states that the strüverite of South Dakota is opaque, but he describes the powder as having a slightly greenish tinge, from which we may perhaps infer that a closer examination of microscopic flakes would reveal optical characters somewhat resembling those of the Perak specimen. In this connection it is interesting to note that the Norwegian ilmenorutile shows a pleochroism closely resembling that of strüverite. Further, as one would expect, flakes of ilmenorutile are more transparent, so much so that flakes showing a good uniaxial figure of positive sign can be obtained without difficulty. Numerous attempts were made to obtain a definite uniaxial figure with the Perak strüverite, but the results were inconclusive, owing to the less transparent nature of the flakes. The optical behaviour of strüverite, however, as outlined above, indicates clearly that it also is positive. (As indicated by compensation with a gypsum-plate, the maximum-absorption vibration-direction is that of the slow ray; and as indicated by the pleochroism it is also the vibration-direction of the extraordinary ray). In view of Prior's suggestion that ilmenorutile and strüverite are solid solutions of mossite and tapiolite respectively, in rutile, it is interesting to note that these minerals are both optically positive, and that the ordinary ray in each case is less absorbed than the extraordinary. In these respects they resemble rutile. Chemical Analysis (S. J. J.).* A preliminary analysis of the mineral showed that it is completely soluble in 5 per cent sulphuric acid after fusion with potassium hydrogen sulphate, and almost completely soluble in concentrated sulphuric acid after digesting for several days. Titanium, tantalum, and iron appeared to be the chief ingredients, but tin, niobium, and silica were also present. Calcium, magnesium, aluminium, chromium, uranium, tungsten, vanadium, zirconium, cerium and its allies, and thorium were proved to be absent. By quantitative analysis the results given in Table I. were obtained. The analysis was made as follows:-The mineral was finely ground and dissolved by heating for several days with hot concentrated sulphuric acid, allowing to cool, and pouring into a large bulk of water. There remained only a slight amount of insoluble residue, which consisted of stannic oxide and silica. A more rapid method of solution My thanks are due to Mr. J. Shelton, A.I.C., Assistant in the Scientific and Technical Department of the Imperial Institute, for valuable assistance rendered in connection with the analysis of strüverite.-S. J. J. The insoluble residue was reduced in hydrogen, and the tin and silica separated in the usual way. The sulphuric acid solution was diluted, nearly neutralised with dilute ammonia, and acidulated with hydrochloric acid, and the tin precipitated by passing a current of hydrogen sulphide for some time. The precipitated stannic sulphide contained a small amount of titanium, from which it was freed by treatment with ammonium sulphide and reprecipitation with acid. The matter insoluble in ammonium sulphide was fused with potassium hydrogen sulphate and added to the hydrogen sulphide filtrate. The stannic sulphide precipitate, together with that obtained from the insoluble residue, was oxidised with nitric acid, ignited, and weighed as stannic oxide. The quantities of stannic oxide found in two determinations by this method were 2.63 and 2'71 per cent. A check determination of the tin, made by fusing the mineral with potassium hydrogen sulphate, dissolving in 5 per cent sulphuric acid, adding a few grms. of tartaric acid, and passing hydrogen sulphide, gave 2.68 per cent of stannic oxide. The filtrates from the hydrogen sulphide precipitation were nearly neutralised, a considerable excess of sodium thiosulphate added, and the whole boiled for about twenty minutes. The precipitate, which contained all the titanium, tantalum, and niobium, together with traces of iron, was freed from the latter impurity by dissolving in fused potassium hydrogen sulphate and re-precipitation as thiosulphate. The iron was obtained from the filtrates by boiling with excess of ammonia, filtering, and re-precipitating as acetate and finally as hydroxide. The iron was weighed as ferric oxide, the gravimetric determination being checked volumetrically by titration with potassium permanganate. Several methods were tried for the separation of tantalic and niobic oxides from the titanium dioxide. Amongst these may be mentioned the following: (a) The fractional crystallisation of the acid potassium fluorides (Marignac method). This was found to be unsatisfactory, owing to the complicating influence of the titanium salt, which interferes with the separation of the tantalic and niobic salts. (b) Fusion with potassium carbonate at a high temperature was also tried, but although the chief part of the tantalic and niobic oxides was soluble, varying amounts of titanium also passed into solution. The addition of potassium nitrate to the melt seemed to intensify rather than diminish this difficulty. The following was found to be the most satisfactory method to adopt. The well-washed thiosulphate precipitate was ignited and ground with about its own weight of pure sugar-carbon, transferred to a small porcelain boat, and heated to a high temperature in a Jena-glass tube in a current of chlorine which had passed through carbon tetrachloride. The temperature of the tube, excluding the portion containing the boat, was maintained at about 70° C., and by this means the less volatile tantalum and niobium chlorides were condensed practically free from titanium. The titanium chloride passed over and was caught by a series of wash-bottles containing water; it was CHEMICAL NEWS, Struverite from the Federated Malay States. July 5, 1912 found impossible to condense the titanium chloride quite | completely by these wash-bottles. By repeating the distillation twice on the more volatile portion, almost complete separation of the tantalum and niobium from the titanium could be effected. The first distillation separated 93 per cent of the total tantalic and niobic oxides found. The trace of manganese present in the mineral remained in the boat. The chlorides of tantalum and niobium were removed from the tube with the aid of strong hydrochloric acid, precipitated as hydrates, washed, ignited, and weighed. This method is essentially similar to that adopted by Wells in separating titanium from tantalum and niobium. The use of carbon tetrachloride alone, as suggested by Wells, was found to be less satisfactory than the method we adopted. A qualitative test for niobium in the above precipitate by the Giles method indicated that this element was present in small quantity (W. B. Giles, CHEMICAL NEWS, 1907, vol. xcv., p. 37). An estimation of the niobium by the Metzger-Taylor method of reduction with zinc and titration with potassium permanganate gave 69 per cent of niobic oxide (F. J. Metzger and C. E. Taylor, School of Mines Quarterly, New York, 1909, vol. xxx., p. 323). The factor used was I cc. N/10 KMnO4-000708 grm. Nb2O5. The tantalum and niobium of another portion of the mixed oxides, which had been separated from the titanium as above, were separated by the Marignac method. By this means the quantity of niobic oxide found was 7.2 per cent. This result is probably slightly in excess of the quantity actually present, owing to the temperature of the experi ment being rather high. The chlorides which had passed over into the washbottles (chiefly titanium chloride) were precipitated by thiosulphate, fused with potassium hydrogen sulphate, the melt dissolved in 5 per cent sulphuric acid, and the titanium and any tantalum and niobium present estimated gravimetrically as oxide. The titanium was estimated volumetrically by the following process:-An aliquot part of the sulphuric acid solution was reduced with zinc and then titrated against N/20 ferric alum solution in an atmosphere of carbon dioxide, using ammonium sulphocyanide as an indicator. The ferric alum solution had been previously standardised against pure titanium sulphate. The results obtained by the two methods showed that the solution obtained in the wash-bottles contained a small percentage of tantalum and niobium, which varied in amount but did not exceed 2 per cent. In the analysis recorded this difference amounted to o'98 per cent, and this quantity was divided proportionately between the niobic and tantalic oxides already found. A direct estimation of the titanium was made by fusing a fresh portion of the mineral in potassium hydrogen sul phate, and, after removing the stannic oxide, estimating the titanium volumetrically by the process given above. An allowance was made for the niobium already found to be present. The titanium dioxide found in this way amounted to 45'74 per cent. The ferrous iron present was estimated by digesting the finely-ground mineral with 50 per cent sulphuric acid in a sealed tube at 200° C. for several weeks and titrating with standard potassium permanganate. The Presence of Scandium in Strüverite.-Prof. A. Fowler, F.R.S., kindly undertook to examine the mineral spectroscopically, and found that it yielded the spectrum of scandium. This is of considerable interest, in view of the rarity of that element, and the fact that strüverite has some features in common with "wiikite." The latter was found by Sir William Crookes to contain more scandia (1∙17 per cent) than any other mineral, and had tantalic oxide, titanium dioxide, ferrous oxide, and silica as its chief constituents (Phil. Trans. Roy. Soc. London, 1908, Ser. A, vol. ccix., p. 17). Since, however, no other mineral examined by him was found to contain as much as o'or per cent of scandium, there seemed to be little chance of finding this constituent in strüverite with the comparatively small amount of material available for analysis; and the 7 chemical examination of the mineral for this purpose led only to the indication of a possible trace. (NOTE. A specimen of orthite has since been described as containing 0.8 to 10 per cent of scandia; see R. J. Meyer, "Ueber einen scandiumreichen Orthit aus Finnland und den Vorgang seiner Verwitterung," Sitzungsber. Akad. Wiss. Berlin, 1911, pp. 379-384. Still more recently a new mineral supposed to consist essentially of scandium silicate, and to contain about 37 per cent of scandia, has been described; see J. Schetelig, "Ueber Thortveitit, ein neues Mineral, Centralblatt Min., 1911, pp. 721-726). Conclusions. (loc. cit.), has now been found in three widely separated Italy; (2) in the Etta mine, Black Hills, South Dakota, localities, viz. (1) at Craveggia, in northern Piedmont, U.S.A.; (3) Perak, Malay Peninsula, and it is rather remarkable, in view of the probable nature of the mineral, that the composition, as seen from the analyses given in Table II., should show such slight variations. The mineral strüverite, as defined by Prior and Zambonini TiO2 TABLE II. South Dakota (Prior). Perak (Johnstone). 41'20 Ta205 Nb2O5 46′96 (a) { 34.8 35'96 (a) This percentage was divided equally between the niobic and tantalic oxides to accord with a rough indication that these two oxides were present in approximately equal amounts. It seems highly probable, however (see below), that there was a substantial preponderance of tantalic oxide. There is apparently no reasonable alternative to the view suggested by Prior, that strüverite is a homogeneous isomorphous mixture, and that it is to be regarded as consisting essentially of a solid solution of tapiolite in rutile. On this assumption we may regard the ferrous oxide of the Perak specimen as combined with the niobic and tantalic oxides to form tapiolite, and treat the mineral as an isomorphous mixture of rutile and tapiolite with small admixtures of cassiterite, silica, and water. (The stannic oxide in the Perak specimen as analysed is present, in part at least, as free cassiterite, and the silica as quartz granules). If now we calculate the separate volumes of these constituents, add these together and divide into the total mass, we get the value 5.33 for the density of the mineral as against the value 5.30 actually found. (Assuming the following specific gravities-rutile 42, tapiolite 7:35, cassiterite 7'o, quartz 2.65, water 1). Treating the analysis of the South Dakota specimen in the same way we get the value of 5.17 for the calculated density of the mineral as against the value 5.25 actually found. The approximation of the calculated to the observed values for the specific gravity may be regarded as sufficiently close, in such a mineral, to be consistent with the solid-solution view. On the same assumption a consideration of the analysis of the Piedmont strüverite in relation to its specific gravity leads to an important conclusion. The Piedmont specimen, as the analysis shows, was much purer than those from South Dakota and Perak. The state of combination of the lime and magnesia is uncertain, but the amount is very small. Neglecting these and recalculating the analysis to 100 in terms of TiO2 and Fe(Ta, Nb)2O6 we get : |