Evolution of Characters and Modularity

In a review of biological usage of the word "character", Colless [4] classi es three di erent types of usage. The rst kind, which he calls an attribute, denotes "a distinctive attribute, quality, or property of structure, form, material, or function". The second kind, which he calls feature, denotes "a part, often of a more or less physical nature, but generally . . .more abstract". The nal kind, which he calls a character variable, denotes "a set of mutually exclusive attributes constituting a logical or mathematical variable". This rather innocent word "character" seems to be su ciently slippery in our minds as to require some critical examination (and to generate this volume). In the somewhat enchiridion like "