EXPERIMENTS ON SURFACE-FRICTION, 251 
Length of surface, or distance from cutwater, in feet. 
2 feet. 8 feet. 20 feet. 50 feet. 
me, G. | 0. tae foes | Col As! B.C. AL | Bal C 
IVEIMALD, ©. coos coseneaenss | 2°00} *A1 |°390 || 1°85/-325 |-264 || 1-8 5/-278 |-240 || 1-83):250 |-226 
Paraffine ...0...5 1.03. 1°95] °388 |°370 || 1°94)°314 |-260 |] 1°93/'271 |-287 |! ... |... |... 
MEEREQUG So cdsostisc cence 2°16} °30 |:295 || 1°99|-278 |-263 || 1-90)262 |-244 || 1°83/°246 |-232 
@aligg: oi <cesseeesscdeinss| 1°93| °87 |°725 || 1-92)°626 |-504 || 1-89)-531 |-447 || 1°87)-474 |-423 
HATE) BAUIC:. cs ccdasca93 > 0% 2°00) ‘81 |690 || 2°00)*583 |-450 || 2°00)480 |-384 || 2°06)-405 |-337 
Medium sand ......... 2°00, ‘90 |°730 || 2°00)°625 |*488 || 2°00!534 |-465 || 2°00|-488 |-456 
Coarse sand ............ | 2°00}1°10 |°880 || 2°00)°714 |-520 || 2°00)"588 |-490 || ... |... |... 
Looking at the subject in its practical aspect, the results exhibited in the 
diagrams and tabular statement may be regarded as literal facts, ascertained 
with great care and exactness by reiterated experiments, the close mutual 
accordance of which was instanced and sufficiently attested by the diagrams in 
plate 4 in the series which accompanied the former Report, in which the 
points deduced immediately from the experiments are shown in connexion with 
the ‘fair lines” drawn through them ; and no difficulty deserving of notice pre- 
sents itself in reference to the practical employment of the results, except 
that, when the probable resistance of a more or less rough surface is to be 
estimated, discrimination must be exercised in selecting, among the qualities 
of surface used in the experiments, that which best serves the purpose of the 
intended comparison. 
Looking at the subject in a speculative aspect, however, certain features of 
the results present perplexing anomalies. 
It is true that the tabulated powers for each quality are, as may be seen, 
very nearly the same, whatever be the length of the surface, presenting only a 
slight tendency to a decrease in the “power” as the length is greater; and this 
difference is not unsuggestive. And again, if in each case, taking the 
resistance at 600 feet per minute as a basis, the resistances at other speeds 
be calculated from this according to the tabulated power, they will be 
found almost in every case to agree very closely, throughout the entire 
line, with those shown in the diagram; and this to a singular degreo as 
regards what is treated as the surface of standard quality—namely, the 
varnished surface. 
But the regularity here exhibited gives additional weight to the discre- 
pancies which appear in other aspects of the effect of quality of surface, and 
some of these seem extremely anomalous; for whereas on comparing the sur- 
faces of tinfoil and again that of scraped paraffine, both of them extremely 
smooth, with the slightly rougher and, consequently, more resisting varnished 
surface, we find that the rougher surface follows the lower power of the speed 
—the power being 2-0 for the tinfoil, 1-94 for the paraffine, and 1°85 for 
the varnish ; we find, on the contrary, in the comparison between the compa- 
ratively smooth varnished surface and the far rougher and far more resisting 
surfaces of calico and sand, that the rougher surface follows the highzr instead 
of the lower power of the speed, the power being 1°85 for the varnish, and 
1-93 and 2-00 (in one case 2:06) for the calico and sand respectively. 
The case of the tinfoil is very remarkable: with a very short plane its 
resistance is little more than half of that of the varnished surface ; yet, possibly 
