JuNE 14, 1901.] 
ity of the bar, weak on account of the relative 
lack of deposition, and weakened by the trans- 
verse scour of the Gulf current, will invite the 
strengthening river-current of the next flood- 
time to break through the bar at that end. In 
this way, there will be a transference of the 
river axis, year by year, toward the left. In 
the meantime the delta has necessarily grown 
most rapidly to the right of the river-mouth. 
The same phenomenon appears to be repre- 
sented in the Rhéne and the Ebro (Figs. 2 and 3). 
Fic. 2. The delta of the Rhone. 
Fig. 1. 
Symbols as in 
In each there are a pronounced leftward deflec- 
tion of the river axis and a corresponding asym- 
metry of delta, coupled witha prevailing marine 
current sweeping past the river mouth from 
left to right. In all three cases, we have de- 
partures from the usual scheme of deflection, 
where the axis of the river is directed down 
stream with respect to the marine current. The 
conditions for this exceptional behavior are: (1) 
a powerful river characterized by astable chan- 
nel, and a delta growing so rapidly as to pre- 
serve one or more distributing arms; (2) a 
nearly tideless receiving water-body with a rel- 
atively steady current transverse to the river 
axis; (8) a’ volume of river sediment greater 
than that of the shore-waste migrating toward 
the delta under the impulse of the littoral cur- 
rent. 
The shape of the bottom, the feeble tidal cur- 
rents and the influence of mud-lumps prob- 
ably have a small effect on the shape of the 
delta as a whole, but no other explanation is 
doubtless so weighty as that found in the force 
SCIENCE. 
953 
of the earth’s rotation. In the northern hem- 
isphere, it tends to produce left-hand deflection 
of an aggrading river. Itis true that the rel- 
Fic. 3. The delta of the Ebro. Cross-lining in- 
dicates older land. Broken lines indicate alluvium. 
The arrow shows the general direction of the marine 
current prevailing at the river mouth. 
ative straightness of river distributaries would 
permit of but a small proportion of the deflect- 
ive force of rotation as affecting a meandering 
stream ; but, small as itis, this force may be 
competent to produce strong asymmetry of a 
delta, since the friction of water against water 
is of a low order. It happens that in the three 
cases above noted, the marine current runsin a 
direction which would control the delta-build- 
ing in the same sense as that expected from 
the influence of the earth’s rotation. What is 
needed, among other tests of the current hy- 
pothesis, is a set of examples where the deflec- 
tion of the river and delta is in appropriate 
relation to the corresponding currents, but in a 
sense opposed to that expected as a result of 
the earth’s rotation. One purpose in outlining 
the hypothesis here in its present brief form is 
to invite observation on this point. Another 
obvious test is experiment. Some rough trials 
with artificial deltas, made on the tidal flats of 
the Annapolis Basin, Nova Scotia, seemed to 
confirm the hypothesis, but other and more 
thorough experiments are needed. Whatever 
be the explanation finally arrived at, it seems 
highly probable that this repeated occurrence 
