58 ECONOMICAL CARGO SHIPS—SOME MODEL EXPERIMENTS. 
values of speed-length ratio. Professor Chapman will see, when he investigates further, 
that there is a very distinct advantage, that he cannot, as he thinks he can, deal with the 
frictional horse-power and then have it out of the way altogether. If he takes a model 
of 75 per cent prismatic coefficient, he will find that by varying the position of the parallel 
middle body he varies the wetted surfaces, and, therefore, after choosing a good parallel 
middle body position and working out his residual resistance on that, he gets back and 
finds he has so varied the wetted surface, it is not the best. With the use of C values you 
can see at a glance exactly the best value for the whole ship, and, what is more, get at 
the results in much less time, and hence the C values have been presented to you. 
Of course Dr. Sadler and others of us know that it is not the simplest thing to plot 
the C values from pounds resistance curves. There are certain corrections to be made, 
dependent on the size of the model and size of the full-sized ship. Once that is done, 
the work is practically all done, because the work of correction in passing from 100 to a 
400 or 800-foot ship is very small and can be laid out as a simple curve as is plotted at 
the foot of Fig. 2, Plate 16, and the curve X also shown on that plate allows you at once, 
with the formula on the edge, to arrive at the exact effective horse-power for the model 
you are dealing with. 
Mr. G. S. Baker, Superintendent, Teddington Tank, England (Communicated) -— 
There are three points which arise in the paper which touch all experimental tanks. 
First we owe our thanks to Mr. Robertson for the data which he has given which 
considerably increases our knowledge in certain directions and will undoubtedly be 
useful for design work. Secondly, with regard to the correction of skin friction in pass- 
ing from model to ship. I am a little lost as to what is actually happening in America 
in connection with this important matter. So far as I am aware there is in existence 
only one genuine set of experiments on the friction of planks in water. There were 
made by W. Froude. They were corrected by his son and amplified by experiments 
made by Gaber and at the Teddington and Washington tanks. These experimental 
results are used at the Teddington tank for all skin-friction correction work. 
No experiments on skin friction were ever made by Tideman, and the statement 
which appears on page 46 of the paper that the method of correcting is by the use of Tide- 
man’s coefficients can only mean that the coefficients used are those adopted by Ann 
Arbor taken from those given by Tideman, which purport to represent Froude’s results 
in another language. It would seem so much more preferable to use, as we do at Ted- 
dington, the original experiments of Froude for one’s correction data than to use this 
longer and necessarily less accurate method. 
With regard to the comparison between tests of the same form made in different 
tanks, we were able through Mr. Luke to test a model at Teddington which had already 
been tried at Washington, and some years later through Admiral Taylor a second form 
was tested in the two tanks. In both cases, when corrections had been made for tem- 
perature and length differences, the results of the two tanks agreed very closely over all 
the useful speed range. Comparison was here made between the ordinary Washington 
varnished wood model and the Teddington wax model. The wax model was afterwards 
coated with varnish and with red lead paint, but these coatings did not have any ap- 
preciable effect on the result. 
One other point touched on by Mr. Robertson in connection with skin friction is 
the effect of fullness of form on skin friction. In the latest edition of my book a method 
