June 28, 1907] 



SCIENCE 



1003 



various ether vapors at 100° C. were obtained 

 by the well-known transpiration method. In 

 the apparatus used the capillary was perfectly 

 straight and the driving pressure obtained by 

 a column of mercury descending under grav- 

 ity. The most interesting substances exam- 

 ined were eight ethers, some of them extremely 

 rare, divided into three groups of isomers. 

 The results show the same fact for these pro- 

 pyl compounds that was observed by Lothar 

 Meyer^ and Steudel for butyl compounds, viz., 

 the molecules of a tertiary compound are 

 smaller than those of a secondary, which in 

 turn are smaller than those of a primary. 



William Campbell read a paper on the iron 

 carbon series of alloys. The various published 

 equilibrium curves of the series, by Roberts- 

 Austen, Rooseboom, Le Chatelier, Benedicks 

 and others, were reviewed. A series of lan- 

 tern slides showed the various changes of 

 structure which take place (a) by variation in 

 composition; (6) by annealing at different 

 temperatures. Two systems were shown to 

 occur: I. austenite (mixed crystals) and 

 cementite; II. austenite and graphite. The 

 former is unstable, the latter stable. 



William Campbell, 



Secretary 



Columbia Untveesitt 



DISCUSSION AND CORRESPONDENCE 



THE CLOCK OF THE U. S. NAVAL OBSERVATORY 



To THE Editor of Science : While not desir- 

 ing to appear to enter into any controversy 

 with the author of the article on ' The Clocks 

 of the Greenwich and the TJ. S. Naval Ob- 

 servatories ' which appeared in your issue of 

 May 31, it would seem that certain facts 

 should be stated to clear up the misunder- 

 standing that has occurred. 



I think no one will disagree with the state- 

 ment that the value of an astronomical clock 

 is to be measured by the degree of accuracy 

 with which its correction can be predicted 

 from observed corrections or interpolated be- 

 tween those corrections : If a series of clock 

 rates extending over several months can be 

 shown to follow such a simple law as that 



'^Pogg. Ann., 1882, Vol. 16, p. 394. 



given on page 451 of Science for March 22, 

 1907, for the Naval Observatory clock, viz. : 



Daily rate = 4- 0^0161 — 0^00103 (T — Mar. 29.0) 

 — 0=.0456 (t — 27°.0), 



and when both these terms have such a prob- 

 able explanation in physical phenomena, it 

 would be folly to refrain from the use of this 

 formula in investigating the running of the 

 clock. It follows that the mean residual 

 O'^.OIS is what really indicates the perform- 

 ance of the clock and not 0^^.035 as deduced by 

 Mr. Lewis. 



The statement by Mr. Lewis that in my 

 article " the Greenwich clock rates are spread 

 over a period of one year " is somewhat mis- 

 leading as they were divided into twelve 

 monthly groups and each group was con- 

 sidered by itself as is clearly shown on page 

 450 of Science for March 22, 1907. That 

 would seem to be as fair a method of treating 

 them as the published data would provide. 



In conclusion attention may be called to 

 the article in Science for April 12, 1907, page 

 570, 'A Eiefler Clock and a Self-registering 

 Right Ascension Micrometer,' in which it has 

 seemed to the writer that the Naval Observa- 

 tory clock runs even better than was indicated 

 by the mean residual 0^.015. 



W. S. ElCHELBERGER 



U. S. Naval Obseevatoey 



VARIATION IN THE COROLLA OF LINARIA 

 VULGARIS MILL 



To THE Editor of Science: In examining 

 the Linaria vulgaris Mill., with a class in 

 botany I found the following remarkable 

 variations in the corolla which may be of 

 interest to some of your readers. The flowers 

 in which the variations appeared were all on 

 the one specimen. 



In the corolla of two of the flowers in which 

 the variations occurred the spur was absent, 

 as was also the usual orange-colored palate. 

 The corolla in both these flowers consisted 

 of five petals, but in one case there were four 

 petals in the upper lip and one in the lower, 

 while in the other all five petals were in the 

 position usually occupied by the upper lip. 



The corolla of a third flower "was tubular. 



