1904-5.] Magnetic Quality in Molecular Assemblages. 1031 
while R takes the place of B as previously used. He then 
calculates by this formula the values of R corresponding to two 
pairs of values of T and Q, as observed by Weiss. The calculated 
values were 323 and 403; Weiss’s observed values were 325 and 
404. I have calculated other values. The numbers in the first 
row below give the field ; those in the second and third rows 
respectively give calculated and observed values of R, the data 
being taken from Weiss’s diagram reproduced in fig. 1. 
10 20 40 60 80 100 120 140 160 200 240 280 320 360 400 440 
91 159 250 299 330 351 363 374 381 389 395 400 404 407 411 414 
100 166 252 299 330 352 364 373 380 388 395 400 404 406 407 409 
Ft§.4. Dissque de tutlc inelaliique. Aimantation pamlUMe el porpendtculatre 
tttt tdiftinp. 
The correspondence is practically perfect for values of the field 
ranging from 50 to 360 ; while, outside these limits, the difference 
is relatively considerable. Weiss’s data are for one specimen only, 
which, as fig. 2 shows, obviously possesses defects ; and further 
■data are very desirable. The peculiar form of the curve Q suggests 
the existence of fissures in planes parallel to the faces of the cube. 
Weiss essentially, and purposely, introduces these in his theoretical 
model. On any view, the partial identity of results is remarkable. 
Wallerant further shows that his expressions lead to the value 
B' = R + R&(cos 2 a cos 2a + cos 2 /? cos 2/3 + cos 2 y cos 2y) 
