This paper is intended as a contribution to the sociology of skill. Re
search which suggests that skills and their transmission are the prope
rties of communities leaves unanswered the question of how information
may be explicitly transmitted and acquired as part of the process of
leaning a skill. Second-order studies of skill accept that skill acqui
sition occurs within a culture, but then go on to examine in detail wh
ich aspects of skills can be explicated and which cannot. Such a secon
d-order study is presented here. Observations of veterinary surgery ar
e used to identify a quasi-quantitative measure of skill acquisition -
hardness. This measure is useful in understanding how task uncertaint
y is resolved in practice and how new skills are learnt.