ane > # 
FEBRUARY 33 1923) Ss 
_ The notice by E. H.-A, and A. E. of Cushman’s 
_“Shallow-water Foraminifera of the Tortugas 
Region”’ (Nature, June 3, p. 708) is timely, for 
there point out, “ It a to be undesirable 
to complicate synonymies the revival of early 
names.’’ They also deprecate the resuscitation of 
-“Discorbis"’ for “ Di ina,” and ‘ Quinque- 
loculina ” and “ Triloculina ” for ‘‘ Miliolina.’’ These 
minor differences in the plans of growth are not 
eneric, for we find them, as often as not, slipping 
‘past the boundaries we have set for them. 
Names of genera, especi those of the mollusca, 
are out-of-date in textbooks almost before they 
; h the hands of our students. Thus Pleurotoma 
of Lamarck was changed, after many years of usage, 
to Turris of Bolten, through the unfortunate discovery 
of a catalogue in which genera were denoted by a 
known species, Turris babylonica, which happens to 
be first on the list. As a case in point, the student 
ets familiar with Turris, but in a few months the 
: her has to inform him that Turris is not only 
extremely restricted but unrepresented in Australia, 
nd the genus has been split up, not into subgenera, 
but into many new genera, useum labels to the 
mumber of several hundreds have to be rewritten, 
and almost before the ink is dry another change may 
be made. 
_ The rule of priority is a good one within bounds, 
but should there not be a retrospective limit placed 
on many groups, dating say from the time when 
‘they were first written upon with authority ? This 
lim om cet might well be settled by a conference 
bf workers in those particular groups. In some 
instances this has been done, and flagrant offences 
against reason have been prevented. Thus, in 1916 
motion for the suspension of the rule in regard to 
the genera Holothuria and Physalia was passed by 
an American conference. In one case “ Holothuria ” 
fas the name given in 1758 to the ‘‘ Portuguese 
Man-of-War,”’ and later, in its familiar sense, to the 
‘Béche-de-Mer, by Bruguiére in 1791. According to 
the rule, “ Holothuria or ‘‘ Béche-de-Mer’”’ being 
i id was to be superseded by “ Physalia,” the 
Mame accepted previously for the ‘ Portuguese 
oon, as olothuria "’ would have become 
ia’ of Jaeger, 1833, and ‘“‘ Physalia ’’ would 
have become ‘ Holothuria ’’ ! 
Even the indispensable and invaluable ‘‘ Index 
alium ’’ of Chas. D. Sherborn will not entirely 
move our troubles, for doubts will arise as to an 
thor’s meaning on account of bad figures and 
escriptions. It is, therefore, of paramount im- 
bortance that a consensus of opinion be obtained 
ar each group as to specific limitations and interpreta- 
ions of authors’ names, and thus prevent those 
eelings of despair which overtake the specialist, and 
@ especially the general worker, at the present 
im. F. CHAPMAN. 
_ National Museum, Melbourne. 






































‘Selective Interruption of Molecular Oscillation. 
‘In Nature of July 22, 1922, vol. 110, p. 112, 
5 mdence occurred regarding the possibility 
ectively interrupting haphazard molecular 
scillation by means of special apparatus, narrower 
certain specific directions than the mean free path 
of the gas in which it was immersed. In view of the 
fact that such methods have now been independently 
ut by Mr. H. H. Platt in America (U.S.A. 
Patent, 1,414,595), the following aspect of the problem 
nay be of interest, particularly since the possibility 
ould appear to be rendered very much more clear 
so regarding it. Fig. 1 represents a portion of a 
NO. 2779, VOL. 111] 
eESpO 

NATURE 

149 
cone longer than that previously considered, its 
diameter, however, still being considerably less than 
the mean free path of the gas concerned, so that 
molecules of the latter may frequently cross from 
side to side without intermolecular interruption. 
Y 



Cc 


5 x 
Fic. 1. 
Let O be any little circle (or sphere if three 
dimensions are being considered) in this cone, and 
consider those molecules proceeding from collision, 
necessarily with equal probability of motion in all 
directions, outwards from the circle. If BC be drawn 
through the centre of O, ea to the top and base 
of the cone, and if AD and XY be drawn equidistant 
from BC, then, provided i ng reflection be presumed 
to occur as an average effect (compare Phil. Mag., 
1922, 43, 1954), it will readily be seen that of the 
molecules issuing from collision in circle O in any 
representative period of time, the ratio of the number 
of those crossing XY to the number of those crossing 
AD, however far (within free path distances) from 
BC these lines may be situated, will always be very 
considerably greater than the ratio of the length XY 
to AD (Phil. Mag., loc. cit. p. 1052). 
If the gas is assumed to be initially of the same 
concentration throughout, and two dimensions only 
are being considered, then the number of molecules 
crossing these lines in any representative period of 
time will either be proportional to their lengths, or 
a change of concentration must occur. It has been 
shown that of the molecules proceeding from collision 
in circle O, an undue proportion will cross XY as 
compared with AD, and this is true (1) for all relative 
sitions of AD and XY within free i distances 
as O, (2) for any and every position of O in the cone. 
It follows, therefore, that molecules starting, with 
equal probability of motion in all directions, from 
collisions in the cone, will create a “‘ condensation ” 
or a disturbance of concentration towards the wider 
rtion of the cone. The same effect may obviously 
e proved fully for three dimensions in a similar 
manner, and is really identical with that dealt with 
in the paper to which reference has been made, since 
Fis. 2. 
the whole cone is merely an extension of the one 
there described, sections ABCD and CDXY both 
being identical with the figure ABCD of the original 
aper. 
3 Eiticequent intermolecular collision in the cone 
cannot destroy the excessive downward bias so 
created, since this will merely be transferred to the 
other molecules concerned. 
Fortuitous rebound from the walls, instead of 
regular ‘‘ reflection,’” may be shown to lead to the 
same effect. If the wall AB be presumed to be ideally 
smooth, then a molecule approaching along path XP 
(Fig. 2) will be ‘‘ reflected ” along PY, receiving an 
impulse from the wall in the direction oc, the wall 
itself receiving the equal and opposite impulse in the 
direction od. If the wall be irregular, or owing to 
E2 
