yime 27, 1872] 



NATURE 



161 



so as the observers were not prepared for an accurate observa- 

 tion of the phenomenon. 



Thus it appears, without anticipating more accurate observa- 

 tions on the occasion of late eclipses, that the shadow-lines were 

 situated in planes perpendicular to the sun's rays. They moved 

 /;'(>w the sun. 



Singularly enough, neither at Tjilentap nor at the island 

 Lawoengan, was anything of tlie phenomenon seen. At the 

 island circumstances were very unfavourable, but at Tjilentap the 

 sky was clear. Oudemans 



liatavia, April 2S 



The Great Storm of June 18 



Out of a large quantity of hailstones collected here after the 



storm had subsided, and which were there'^ore partly melted, 1 



selected one of the largest. The subjoined boundary line is the 



measure of a section of this hailstone through the poles, the 



form of it being a prolate spheroid, llattened on two sides, like 

 a confectionary lozenge, if I may employ so vulgar a comparison. 

 Many of the hailstones, however, as they fell, were jagged 

 pi' ces of ice, the like of wliich I have never seen. 



Kdgbaston, Birmingham, June 21 C. M. Inglebv 



Spectrum of Lightning 



I HAD a good view of the spectra of lightning during the storm 

 of yesterday. Frequently there was only one bright line visible, 

 this being coincident with the nitrogen hue. At other times there 

 were several bright lines, sometimes with, and at other times 

 without, the nitrogen line. Several flashes showed a continuous 

 sprectrum without visible lines. My instrument was a small 

 direct-vision spectroscope, but sufficiently powerful to divide the 

 sodium line. J. P. Joule 



Broughton, Manchester, June ig 



YESTliRDAY this neighbourhood was visi'ed by a most terrific 

 thunderstorm, such as I have never before seen in England. 

 Indeed, it is stated that the last storm of similar severity occurred 

 exactly }i}, years ago to a day — rather a singular coincidence. 

 The storm commenced here about half-past one, though distant 

 thunder was audible at one o'clock. It was accompanied by 

 violent wind, rain, and hail, and lasted about an hour, during 

 which o 66 in. of rain fell. The wind was S. E. at the time, but 

 the storm came from tlie W. The hailstones are described as 

 being, many of them, larger than marbles, and did a good deal 

 of damage to glass. Several fatal accidents have occurred from 

 the lightning, which for some time was almost incessant. I' 

 examined its spectrum with a miniature spectroscope, and suc- 

 ceeded in observing four or five lines. Their approximate posi- 

 tions, which I give below, are very roughly determined, and 

 especially so toward the red end. . They were obrained by com- 

 parison with the Fraunhofcr lines, and with the carbon spectrum 

 of a Bunsen burner. The former were mostly very faint from 

 the darkness, but the atmospheric absorption bands near D were 

 very marked. I am strongly of opinion that the spectrum is 

 that ordinarily given by a spark in air, but was unable to make 

 direct comparison. I have since examined a feeble spark by the 

 same spectroscope, and the general appearance is very similar. 

 I also tried obser\'ation of the long zig-zag Hashes with a simple 



prism, but without much success, though I was able to see the 



spectrum. H. R. Procter 

 North Shields, June ig 



Lines. Wave-length. Remarks. 



a about 66 Sth-metres — 



3 ,, 59 „ Doubtful. 



7 ,> 56 „ — 



Bright. 



Water Analysis 



Mv attention has been directed to an article entitled " Water 

 Analysis, I." published in Nature of June 6. The article is 

 unsigned, bearing neither name nor initials, and contains strange 

 errors and misrepresentations, some of which I beg permission 

 to correct. 



First, there is a false date. The article states that in 1868 

 " Messrs. Chapman, Wanklyn, and Smith proposed to determine 

 the lUi^'anic matter in water from the amount of ammonia evolved 

 when tlie water was treated with a strongly alkaline solution of 

 potassic permanganate, and then distilled. ' The truth is, that 

 our paper, proposing the process, and giving directions how to 

 work it, together with examples, was read before the Chemical 

 Societv on fune 20, 1867, and publi-hed in the jfoin'>!al for the 

 year 1867 ('c'li/c p. 445, d st'i/ ). Moreover, in the year 1867, our 

 process was extensively employed byjthe Rivers Commission by 

 Mr. Way, who was at that period the chemist on the Commission. 



Next, I have to notice a misrepresentation. The article de- 

 scribes us as having at first stated that albumen gave up the 

 whole of its nitrogen (in the form of ammonia) when treated 

 with alkaline permanganate, and that altc-rwards we said that 

 onlv a certain fraction was obtainable in that way. 



VVe have never said that distillation of albumen with alkaline 

 permanganate converted the whole of the nitrogen of the albu- 

 men into ammonia. The assertion in the article is therefore 

 untrue. The circumstance to which your siatemcnt was intended 

 to refer was the following. 



In our paper read on June 20, 1S67, we proposed two distinct 

 modifications of the water process. In the one modification we 

 evaporated to dryness with potash in the oil-bath, and afterwards 

 distilled the residue with alkaline permanganate. The quantity 

 of ammonia got by the operation with potash in the oil-bath, 

 f/iis the quant ty of ammonia got afterwards by permanganate of 

 potash, is equal or nearly equal to the total ammonia which the 

 total nitrogen of the albumen will yield. 



On June 20, 1867, in addition to this eaily form of the water 

 process, we described and recommended a second modification, 

 consisting iu the omission of the evaporation to dryness with 

 potash. We boiled with potash, but did not take down to dry- 

 ness, and then boiled wiih permanganate. At that date we 

 knew, and mentioned in the paper, that omission to take down 

 to dryness involved some loss of ammonia which potash should 

 evolve. We did not know that failure to get the full yield with 

 potash involved the u't'inate sacrifice of a certain quantity of 

 ammonia. That fact was afterwards ascertained by me, and 

 published later in the autumn of 1S67, and is duly recorded in the 

 Jou! iial of the Cliciiiic.l Society. 



The conviction that a really serviceable process of water- 

 analysis must be a simple one, and the perception that a definite 

 fraction of the total nitrogen was as good a datum as the total 

 nitrogen itself, led me to persist in recommending the second 

 modification rather than the first. Much experience in these 

 matters has confirmed my judgment, and I do not repent the 

 choice that we made. 



Returning to the article. After having mentioned our experi- 

 ments on papaverine, sulphate of cinchonine, narcotine, strych- 

 nine, sul] hate of quinine, there is the following extraordinary 

 statem»nt : — 



" If the authors had enabled us to ascertain the absolute error 

 on the quantity taken instead of the percentage error, by giving 

 us the quantities from which the resul's were taken, it would no 

 doubt be much more apparent : the resuhs given above in the 

 Gate of Frankland and Armstrong's paper are absolute errors." 



I invite you to open \\\e Journal oj the Chemical Soeuty, May 

 1868, which is referred to in our treatise. We did give the quan- 

 tities from which the results were obtained. Quoting from our 

 memoir, you may read that we took 10 mgrm. o( papaverine, and 

 obtainedo'22 mgrm. of ammonia; that we took 10 mgrm. and 5 

 mgrm. of sulphate of cinchonine, and got respectively 0'57 and 



