402 



NA TURE 



[February 23, 1893 



soluble in water and alcohol, the solutions being coloured dark 

 green, and the salt may be recrystallised from these solvents. 

 Upon analysis they were found to consist of the chloride OsjCIy 

 crystallised with seven molecules of water. 



This chloride of osmium, OsjCly.yHoO, would appear to be a 

 molecular compound of the trichloride, OsCla, and the tetra- 

 chloride, OSCI4. For when potassium chloride solution is added to 

 the solution of the crystals in alcohol, a precipitate of brilliant red 

 octahedrons and cubes of potassium osmichloride, K20sC]6, is 

 obtained, showing the presence of osmium tetrachloride, OsClj. 

 Moreover, when the precipitate is separated by filtration, and 

 the filtrate concentrated by evaporation in vacuo, dark green 

 crystals of the trichloride, OsCIj, are deposited containing three 

 molecules of water of crystallisation. 



During the reduction of these crystals of the trichloride in a 

 current of hydrogen for the purposes of analysis, a small quantity 

 of a white sublimate was obtained, which probably consisted 

 of the octo-chloride, OsClg, corresponding to the tetroxide OSO4. 



Bromine does not react with osmium with anything like the 

 energy of chlorine. The free elements do not appear to com- 

 bine at all, even at moderately high temperatures. Only a small 

 quantity of a sublimate of a dark brown colour is obtained by 

 passing bromine vapour over osmic acid. This sublimate dis- 

 solves to a brown solution in water, which, however, rapidly 

 decomposes with deposition of a black precipitate. 



When osmic acid, H2OSO4, is treated with hydrobromic acid 

 in the manner just described in the case of hydrochloric acid a 

 similar reaction occurs with formation of a clear reddish-brown 

 solution which yields, upon evaporation i^t vacuo over sulphuric 

 acid and solid caustic potash, small crystals of a molecular 

 compound of the tribromide, OsBrg, and the hexabromide, OsBrg, 

 together with six molecules of water of crystallisation. These 

 crystals of OsjBrg.eHaO are dark reddish-brown in colour and 

 exhibit a beautiful metallic lustre. They are quite stable when 

 preserved in a dry atmosphere, but rapidly deliquesce in moist 

 air. 



Iodine appears to possess even less affinity for osmium than 

 bromine. When, however, osmic acid is treated with hydriodic 

 acid a deep greenish-brown solution is obtained which deposits 

 in vacuo dark violet rhombohedrons, exhibiting a brilliant 

 metallic lustre, consisting of the anhydrous tetra-iodide of osmium, 

 OSI4. This iodide, the only one containing osmium yet pre- 

 pared, is permanent in a dry atmosphere at the ordinary 

 temperature, but rapidly deliquesces like the bromide when 

 exposed to moist air. 



In relative stability the chloride bromide and iodide of osmium 

 above described exhibit a gradation such as would be expected 

 from the relations between the halogen elements themselves. The 

 iodide is readily dissociated by slightly raising the temperature, 

 and upon the addition of water is decomposed with the deposi- 

 tion of a black precipitate containing the metal. A similar 

 decomposition occurs, although much more slowly, in case of 

 the bromide. The chloride, however, is well-nigh permanent 

 under these conditions, only exhibiting traces of decomposition 

 after the lapse of a considerable time. A. E. Tutton. 



REDUCTION OF TIDAL OBSERVATIONS} 



T^HE tidal oscillation of the ocean may be represented as the 

 -^ sum of a number of simple harmonic waves which go 

 through their periods approximately once, twice, thrice, four 

 times in a mean solar day. But these simple harmonic waves 

 may be regarded as being rigorously diurnal, semi-diurnal, ter- 

 diurnal, and so forth, if the length of the day referred to be 

 adapted to suit the particular wave under consideration. The 

 idea of a series of special scales of time is thus introduced, each 

 time-scale being appropriate to a special tide. For example, 

 the mean interval between successive culminations of the moon 

 is 24h. 50m. , and this interval may be described as the mean lunar 

 day. Now there is a series of tides, bearing the initials Mj, 

 Mj, Mj, M4, &c., which go through their periods rigorously 

 once, twice, thrice, four times, &c., in a mean lunar day. The 

 solar tides, S, proceed according to mean solar time, but, be- 



' " On an Apparatus for facilitating the Reduction of Tidal Observations." 

 By G. H. Darwin, F.R.S., Plumian Professor and Fellow of Trinity 

 College, Cambridge. A paper read before the Royal Society on Decem- 

 ber 15, 1892. 



NO. I 2 17, VOL. 47] 



sides mean lunar and mean solar times, there are other special 

 time scales appropriate to the other tides. 



The process of reduction consists of the determination of the 

 mean height of the water at each of twenty-four special hours, 

 and subsequent harmonic analysis. The means are taken over 

 such periods of time that the influence of all the tides governed 

 by other special times is eliminated. 



The process by which the special hourly heights have hitherto 

 been obtained is the entry of the heights observed at the mean 

 solar hours in a schedule so arranged that each entry falls into 

 a column appropriate to the nearest special hour. Schedules 

 of this kind were prepared by Mr. Roberts for the Indian 

 Government.^ The successive rearrangements for each sort of 

 special time were made by recopying the whole of the observa- 

 tions time after time into a series of appropriate schedules. 

 The mere clerical labour of this work is enormous, and great 

 care. is required to avoid mistakes. 



All this copying might be avoided if the observed heights were 

 written on movable pieces. But a year of observation gives 

 8760 hourly heights, and the orderly sorting and resorting of 

 nearly 9000 pieces of paper or tablets might prove more laborious 

 and more treacherous than recopying the figures. 



The marshallingof movable pieces might, however, be reduced 

 to manageable limits if all the twenty-four observations pertaining 

 to a single mean solar day were moved together, for the mov- 

 able pieces would be at once reduced to 365, and each piece 

 might be of a size convenient to handle. 



The realization of this plan affords the subject of this paper, 

 and it appears that not only is all desirable accuracy attainable, 

 but that the other requisite of such a scheme is satisfied, namely, 

 that the whole computing apparatus shall serve any number of 

 times and for any number of places. 



The first idea which naturally occurred was to have narrow 

 sliding tablets which should be thrown into their places by a 

 number of templates. It is unnecessary to recount all the trials 

 and failures, but it will suffice to say that the slides and tem- 

 plates would require the precision of a mathematical instrument 

 if they are to work satisfactorily, and that the manufacture 

 would be so expensive as to make the price of the instrument 

 prohibitive. 



The idea of making the tablets or strips to slide into their 

 places was accordingly abandoned, and the strips are made with 

 short pins on their under sides, so that they can be stuck on to 

 a drawing board in any desired position. The templates, which 

 were also troublesome to make, are replaced by large sheets of 

 paper with numbered marks on them to show how the strips 

 are to be set. The guide sheet is laid on a drawing board, and 

 the pins on the strips pierce the paper and fix them in their 

 proper positions. 



The strip belonging to each mean solar day is divided by 

 black lines into 24 equal spaces, intended for the entry of the 

 hourly heights of water. The strip is nine inches long by i inch 

 wide, and the divisions (| byi) are of convenient size for the 

 entries. There was much difficulty in discovering a good 

 material, but after various trials artificial ivory, or xylonite, was 

 found to serve the purpose. Xylonite is white, will take writing 

 with Indian ink or pencil, and can easily be cleaned with a damp 

 cloth. It is just as easy to write with liquid Indian ink as with 

 ordinary ink, which must not be used, because it stains the 

 surface. 



The observations are to be treated in groups of two and a 

 half lunations or 74 days, A set of strips, therefore, consists 

 of 74, numbered from o to 73 in small figures on their flat 

 ends. 



If a set be pinned horizontally on a drawing board in vertical 

 column, we have a form consisting of rows for each mean solar 

 day, and columns for each hour. The observed heights of the 

 water are then written on the strips. 



When the twenty-four columns are summed and divided by 

 the number of entries we obtain the mean solar hourly mean 

 heights. The harmonic analysis of these means gives the mean 

 solar tides. But for evaluating the other tides the strips must 

 be rearranged, and to this point we turn our attention. 



Let us consider a special case, that of mean lunar time. A 

 mean lunar hour is about ih. 2m. m.s. time ; hence the I2h. of 

 each m.s. day must lie within 31m. m.s. time of a mean lunar 



I An edition of these computation forms was reprinted by aid of a grant 

 from the Royal Society, and is sold by the Cambridge Scientific Instiument 

 Cc mpany, but only ab(. ut a dozen copies now remain. 



