370 LIFE AT DOWN. ^-TAT. 33-4$. [1849. 



Thompson, who is fierce for the law of priority, and he gave 

 it up in such well-known names. I am in a perfect maze of 

 doubt on nomenclature. In not one large genus of Cirripcdia 

 has any one species been correctly defined ; it is pure guess- 

 work (being guided by range and commonness and habits) to 

 recognise any species : thus I can make out, from plates or 

 descriptions, hardly any of the British sessile cirripedes. I 

 cannot bear to give new names to all the species, and yet I 

 shall perhaps do wrong to attach old names by little better 

 than guess ; I cannot at present tell the least which of two 

 species all writers have meant by the common Anatifera 

 l&vis ; I have, therefore, given that name to the one which is 

 rather the commonest. Literally, not one species is properly 

 defined ; not one naturalist has ever taken the trouble to open 

 the shell of any species to describe it scientifically, and yet all 

 the genera have half-a-dozen synonyms. For argument's sake, 

 suppose I do my work thoroughly well, any one who happens 

 to have the original specimens named, I will say by Chenu, 

 who has figured and named hundreds of species, will be able 

 to upset all my names according to the law of priority (for he 

 may maintain his descriptions are sufficient), do you think it 

 advantageous to science that this should be done : I think 

 not, and that convenience and high merit (here put as mere 

 argument) had better come into some play. The subject is 

 heart-breaking. 



I hope you will occasionally turn in your mind my argument 

 of the evil done by the " mini " attached to specific names ; 

 I can most clearly see the excessive evil it has caused ; in 

 mineralogy I have myself found there is no rage to merely 

 name ; a person does not take up the subject without he 

 intends to work it out, as he knows that his only claim to 

 merit rests on his work being ably done, and has no relation 

 whatever to naming. I give up one point, and grant that 

 reference to first describer's name should be given in all sys- 

 tematic works, but I think something would be gained if a 



