( 1093 ) 
a serpentine shape, Owing to the fact that each bright interference 
fringe, at its intersection with a dark line, seems to push it toward 
the right at the lower edge, toward the left at the upper edge of 
the fringe. This proves that the darkness of those lines is not exelu- 
sively caused by absorption, but partly by anomalous dispersion. 
The serpentine shape of the lines is not an optical illusion, for the 
sodium line D, (D,, which lies at 2'/, m.m. distance to the right, 
is invisible on the reproduction), originating in the are, is sharp and 
perfectly straight. So we may take it for etablished by experiment, 
that thousands of lines occurring in the spectra of iodine vapour, 
bromine vapour, and nitrogen peroxide, produce anomalous dispersion 
of the waves lying close to them. 
The result of this investigation supports the thesis, that selective 
absorption is always accompanied by anomalous dispersion. 
Physics. — “Further experiments with liquid helium.’ By Prof. 
H. KAMERLINGH ONNEs. Communication N°. 119 from the 
Physical Laboratory at Leiden. 
(Communicated in the meetings of Dec. 24, 1910 and February 25, 1911). 
A. TLsotherms of monatomic gases ete. VIL. Thermal 
properties of helium. 
§ 1. The helium-bath. In most of the experiments that one would 
like to make at helium temperatures, it is necessary to transfer the 
liquid helium from the apparatus in which it has been prepared to 
another — the helium cryostat — more suitable for holding the 
apparatus arranged for these special experiments. In Comm. N°. 112 
(June 1909) it was mentioned that this was going to be tried; and 
in the Jubilee book presented to J. M. van BemMMELEN it was further 
stated that the transference had, in fact, been once successfully 
accomplished. Although the suecess which attended this operation 
allowed immersion in the protected helium bath of the apparatus, 
with which it was shown that even at a vapour pressure as low as 
0.15 mm. helium is still a liquid, it was nevertheless evident that 
this desirable result had been obtained only by accident. A method 
that promises to be more efficient is now being developed, and I 
hope to be able to make a communication in the near future 
concerning it. 
In the meantime, a few problems could already be studied with the 
help of a liquefying apparatus resembling the original liquefying 
apparatus (Comm. N°. 108, Proc. of May/June 1908) sufficiently well 
