626 



SCIENCE. 



[N. S. Vol. IX. No. 226. 



using names that have more or less similarity. 

 Its purpose, as stated, was ' to illustrate what 

 the present system is leading to.' Names of 

 formations and dates of publication were given 

 for this purpose. 



However, Director G. M. Dawson, in his re- 

 cent communication in Science, states that it 

 is not apparent from my remarks that Cache 

 Creek group of formation holds priority. I do 

 not see how any other construction can be given 

 to the third paragraph of my letter. It is there 

 briefly shown that Dr. Selwyn described the 

 Upper and Lower Cache Creek group in 1872, 

 and that in 1896 Dawson applied the name 

 Cache Creek formation to both series. It is 

 further evident, from the names and dates given, 

 that the Cache Creek group has priority over 

 either Cache Valley group or Cache Lake beds. 

 F. B. Weeks. 



U. S. Geological Survey, 

 Washington, D. C. 



NOTES ON INORGANIC CHEMISTRY. 

 At the meeting of the Chemical Society (Lon- 

 don) on March 16th Professor Dewar pre- 

 sented a paper on the boiling point of hydrogen, 

 which is printed in the Proceedings. In obtain- 

 ing liquid hydrogen great difficulty is expe- 

 rienced owing to the presence of small traces of 

 air. Quantities amounting to only one thous- 

 andth of one per cent, accumulate in the solid 

 state and eventually choke the nozzle of the ap- 

 paratus, necessitating the abandonment of the 

 operation. Dewar obtained 250 cubic cen- 

 timeters of colorless liquid hydrogen and used 

 this for the determination of the boiling point. 

 His previous observations, using a platinum 

 resistance thermometer, gave the boiling point 

 as — 238°. In these latest experiments a possi- 

 ble constant error in the use of the platinum 

 thermometers was checked by using a rhodium- 

 platinum resistance thermometer, the alloy con- 

 taining ten per cent, rhodium. Examination 

 had shown that alloys, unlike pure metals, 

 showed no sign of becoming perfect conductors 

 at absolute zero. The rhodium-platinum ther- 

 mometer gave the boiling point of hydrogen as 

 ■ — 246°, and this the author considers to be 

 more accurate than the previous determinations, 

 especially as it agrees very fairly with the boil- 



ing point calculated from the results of Wrob- 

 lewski and of Olszewski. 



In an addendum dated March 17th Dewar 

 gives the first results from a constant-volume 

 hydrogen thermometer working under dimi- 

 nished pressure. This gave — 252° as the boil- 

 ing point of hydrogen. The three results in 

 absolute temperature are : (1) platinum resist- 

 ance thermometer, 35°; (2) rhodium-platinum 

 resistance thermometer, 27°; (3) hydrogen ther- 

 mometer, 21°. From this Dewar states that 

 it appears that the boiling point of hydrogen is 

 really lower than was anticipated. 



In the Journal of the Society of Chemical In- 

 dustry the use of titanium compounds in the 

 dyeing industry is discussed, and it is shown 

 that in many cases they may be successfully 

 utilized. Especially are they valuable as 

 mordants for alizarin yellows and oranges, and 

 for basic dyestufls. The tannate is not unstable 

 towards acids and little influenced by light, and 

 according to the author is valuable as a water- 

 color. As experiments progress it is by no 

 means impossible that many of the elements 

 which now have little or no practical value may 

 find uses, and work along this line offers much 

 prospect of success. 



In the last number of the Zeitschrift fiir an- 

 organische Chemie, Piccini publishes the full de- 

 tails of the preparation of cesium manganese 

 alum and a complete description of the salt. 

 This is of more than passing interest from the 

 fact that it is the first salt of trivalent man- 

 ganese whose constitution is not open to 

 question. Manganese alums are described in 

 older chemical literature, but efforts to repeat 

 their preparation have not been successful. By 

 utilizing the electric current to oxidize man- 

 ganese sulfate, Piccini forms the alum with- 

 out difficultj', and in crystals large enough for a 

 complete crystallographic study. It is thus 

 settled that in manganic compounds the man- 

 ganese is trivalent, and hence allied to aluminum, 

 chromium and iron. From a private communi- 

 cation I learn that other manganese alums have 

 been prepared and studied by Professor Chris- 

 tensen, and will shortly be described. 



J. L. H. 



