NOTES ON RIVETS AND SPACING OF RIVETS FOR OIL-TIGHT WORK. 63 
sought out weak spots and loosened rivets developed under new stresses created 
by carrying oil in bulk. 
The first leaks frequently occurred at the ends of stiffeners and in rivets con- 
necting frames to shell which were subject to stresses not common in ordinary 
vessels. After these started the vessel began to work more, and progressive 
damage followed, frequently resulting in almost entirely reriveting the vessel. 
There was a fad about this time for plug-headed rivets, but these were aban- 
doned because they would not draw the work up sufficiently close and created 
greater difficulty in securing high quality of workmanship. The old reliable pan 
head was found to produce the most intimate contact and to be more permanently 
oil-tight and was recommended except where some other types were absolutely 
necessary. 
Up to 1894 there were built or building to Lloyd’s Class sixty-four tankers, 
yet the Society did not consider that it had sufficient experience to lay down 
definite rules for their construction. In this year, however, Mr. Benjamin Martell, 
chief surveyor to Lloyd’s Register, read a paper before the Institution of Naval 
Architects wherein he outlined his views on the details of construction and of 
riveting for bulk oil carriers, recommending three diameters for oil-tight work. 
This paper in pamphlet form was distributed and for a number of years served as 
a guide to shipbuilders in the design and construction of tankers intended to be 
classed at Lloyd’s. 
Within a Comparatively short time after this, Bureau Veritas, apparently 
profiting by this paper, published rules for tankers, and the American Bureau of 
Shipping published their first rules in 1902, but the first rules of Lloyd’s did not. 
appear for general use until 1909. 
Both Bureau Veritas and the American Bureau recommended three and one- 
half diameters at the start, while Lloyd’s specified one more rivet in a frame space 
than required for water-tight spacing resulting approximately in three and one- 
half diameters. 
According to the recollection of the writer there was a period when Lloyd’s 
practically decided not to class any more converted vessels, but both workman- 
ship and details of construction must have improved, as many converted vessels 
have since been classed and have given good service. 
Evolution was a slow process, and no doubt in numerous cases the true cause 
for failure was not recognized. Mr. Martell realized this when he wrote his paper 
and provided for tanks not over 24 feet in length, increased stiffeners and brackets, 
as well as closer spacing of rivets. Had the earlier tankers been better designed as 
regards tanks, stiffeners, brackets and compensation, where needed, the question 
of rivets and spacing of rivets for oil-tight work might possibly have received 
somewhat less attention during those troublous times. This appears to be a rea- 
sonable inference from the fact that so many converted tankers have since given 
long and satisfactory service. 
Rivets in the shells of the early tankers were continually getting loose and 
- leaky, while it is very rare that the shell rivets spaced for water-tight work of later 
