Ogt lute 
a/(r/2) (in addition to being made dimensional on the scale of 
(H/K) Y2 ). Thus, the theory predicts that wave height along the side 
of the wedge will be proportional to the wedge angle. 
Calculation of Z(X) has been carried out, with the results 
shown in Figure 2. In addition, the integral in (18) has very simple 
asymptotic approximations which are validas X 40 or Xo, and 
these are shown by the broken curves in Figure 2, 
In order to determine whether this result was even approxi- 
mately valid, we conducted some experiments with a very simple 
model. The planform of the model was that of an unsymmetrical dia- 
mond ; at one end, the model was a wedge with a half-angle of 7.5°, 
and at the other end the half-angle was 15°. Tests were conducted at 
speeds up to about 15 ft. /sec., with drafts from 4 in. to 16 in. 
A grid had been inscribed on the model so that wave shapes could be 
measured from photographs of the bow wave. 
In Figures 3 and 4, two selected series of tests are shown. 
In both figures, the model is being tested at a draft of 12 in. 
There are several qualitative features in these photographs 
that are worth noting : 
(i) The model speed in Figure 3(a) is 1.64 ft. /sec., which is 
only about twice the minimum speed at which waves can travel.ona 
water/air interface. (Minimum speed is about 23.2 cm./sec.) In 
fact, capillary waves are quite evident in this picture, as well asin 
several of the higher-speed test pictures. Whether these ripples can 
actually be seen apparently depends more on the lighting than on any- 
thing else. The existence of a sharp edge on the model presumably 
accentuated the amplitude of the ripples in all of our tests. 
(ii) In (b) - (e) of Figure 3, the water level at the bow edge is 
about 1 in. above still-water level. (The white mark at the bow is 
atthe 18in. draft mark, and the squares are 1 in. ona side.) This 
rise of water level ahead of the bow is, of course, not predicted in 
the analysis. We fully expected to observe sucha rise, and we re- 
cognized that it would represent a source of error in the predictions, 
What we did not anticipate was that the rise is quite insensitive to 
forward speed. From a speed of about 5 ft. /sec. (Figure 3(b)) toa 
speed in excess of 15 ft. /sec. (Figure 3(e)), this rise increases 
from about 0.8 in. toabout 1.2 in, 
(iii) The corresponding rise in water level at the bow is greater 
for the wider-angle bow, but even in this case the level seems to 
L520 
