(i) Errors in the original recording of or counting the number of ships. Nothing can be done about this; there is no reason to suppose it has much importance, though the existence of a smuggling fleet in the eighteenth century should be borne in mind.

(ii) Inconsistencies in contemporary recording of the tonnage of individual ships. This is much less important than many writers have supposed; almost invariably recorded tonnages are tons burden. Though, an individual ship may be put down at different times at several different tonnages, over a great mass of material these differences should be evened out; there is no reason to think that there were consistent errors in one direction followed at a later period by consistent errors in another. Two exceptions must be mentioned, however. One is the practice of recording the tonnage of East Indiamen, from the early years of the eighteenth century onward, at 499 tons or less, though many of them were far larger. This I shall remark again later. The other is the growth after 1773 of the practice of recording measured tons rather than tons burden; of no importance in the period we are specifically concerned with, it makes comparisons with later dates very difficult - especially as measured tonnage came to be used invariably in records after general registration began in 1786.