East Sails West: The Voyage of the Keying, 1846–1855

10. The Keying’s Dimensions and Shape

The measurements we can glean from the Description more or less fit with all other sources, though since all of these probably came from Captain Kellett, any errors he made have merely been replicated. We know from the Description, for example, that an attempt to measure the Keying in New York caused so much trouble that those who made the attempt threw up their hands in despair and gave up.1 However, even here there is confusion. Table 4 below gives the dimensions from various sources, from which it can be seen that we cannot even be confident of the measurements. Chapter 10 The Keying’s Dimensions and Shape Table 4 The Keying’s dimensions from all known sources Dimensions Description Other Length overall 165' 160' (medal); 165' (Worcester, Wikipedia) Beam 25' 6" 33' (Worcester); 35' (Wikipedia); 35' 4" (Haddad) Depth of hold 12' 16' (Worcester); 19' (Wikipedia, medal) Mainmast 95' 90' (Worcester); 85' from deck (medal) Foremast 75' 75' Mizzenmast c. 50' 50' Mainsail area 9,900 sq. ft. (1,100 sq. yd.) 11,000 sq. ft. (Worcester) Tonnage 400 (measured) 700, 750 or 800 (Worcester, medals) 700 (approx. deadweight) 800 tons (Wikipedia) 800 Chinese measurement (medals) 218 East Sails West The only additional piece of evidence from the Description is that the Keying was a Second- or B-class junk. There is no certain way of deciding what this actually means, since the bracket for B-class junks was quite wide and very hard to tie down in terms of actual size. Paul van Dyke’s excellent The Canton Trade2 gives the rough basis for calculating the length and breadth of an average B-class vessel as being from 15 to 20 measurement covids in beam and from 70 to 80 measurement covids in length. Milburn’s Oriental Commerce3 gives the range for a B-class junkasfrom71to74covidsinlengthandfrom22to23covidsinbeam.Theexact length of a covid is uncertain. It is given in Blunt’s The Shipmaster’s Assistant and Commercial Digest (1837) as 14.625 inches.4 Against that, although in Milburn’s extensive entry for China no value for the covid is given, his entries elsewhere value the covid or ‘China cubit’ as between 16 and 19 inches.5 Using the full range implicit in these various values would make a B-class junk from 24.4 to 36.4 feet in beam and from 85.3 to 127 feet in measured length. Worcester also notes the variability of Chinese linear measure, depending on trade and purpose.6 This is obviously a far smaller value than we should expect for the Keying, but not necessarily a surprise, for two reasons. First, since the ‘measurement covid’ (or cubit) was a fiscal measure and does seem to have varied—though by how much, in which sense and in what years is unfathomable—turning a rough bracket of values into useable data is inevitably difficult. Second, the ‘length on deck’ measure was specified as the distance between the aft side of the foremast and the forward side of mizzen mast—or, according to William Hunter, the forward face of the rudder head7 —so it is necessarily different from length overall. Using the Currier print as a rough guide, the Keying’s foremast was about one-fifth of the length overall abaft the forward end of the ‘wings’; the mizzen mast was about the same distance forward of the transom. However, although the measured length must therefore be adjusted by quite a bit, the beam would not change greatly. Allowing for plank thickness and the protrusion of the main longitudinal wales, but excluding catwalks, the actual beam to the ship’s outside planking would probably be some two to six feet greater than what a B-class junk measured. Given the bracket of values above, that means a beam measurement of between 26.4 and 42.4 feet. This embraces all the beam measurements in Table 4. For the difference between measurement length and length overall, following The Keying’s Dimensions and Shape 219 the proportions shown in the Currier print, the length measurements above should range from 102.4 to 152.4 feet. Here, there is a marked difference to the values in Table 4, the junk seeming to measure as far too short. This could simply be a mistake in the description of the Keying. She may have been an A-class junk, and therefore over 74 covids in measured length...