Evolution and Symmetry in the Order of the Sea-pens. 129 



in which there seem to be distinct specific differences between groups of 

 specimens belonging to the same genus. For example, the two species 

 Alcyonium digitatuvi. and A. glomeratum appear to be quite good species. 

 They are found close together in some parts of the waters of the British 

 coast (e.g., tlie south coast of Cornwall), and the first-named species, which 

 is by far the most abundant, exhibits a wide range of variation in form, 

 colour, and spiculation. Nevertheless, I have not yet found a single variety 

 of A. digitatum among the many hundreds I have examined that could 

 possibly be mistaken for A. glomeratum, or any other species of the genus. 

 But even if it be established that specific discontinuity does occur in some 

 genera of sedentary animals, it does not follow that such discontinuity occurs 

 in all or even in a majority of them, and the evidence, so far as it goes at 

 present, tends to show that it is exceptional rather than universal. 



In attempting to make a scientific classifieation of any group of animals it 

 is found that the characters of the Greatest value are those that are least 

 variable ; but characters that are very variable in one family or genus may 

 be much less variable in another family, and it is therefore necessary, before 

 a suggested scheme of classification can be regarded as stable, to ascertain 

 the range of variation of the characters it is proposed to use. 



Thus in the shrimp Hippolyte the character " colour " is so variable that 

 it cannot be used in specific diagnosis, but in the crayfish Astacus 

 (Potamobius) the red colour of the chela; of Astacus fiaviatilis, and the pale 

 colour of the chelae of A. pallipcs are invariable characters, supported by 

 others, for distinguishing the species. 



In the sea-anemone Metridium marginatum only 33 per cent, of 131 adult 

 specimens exhibited the arrangement of mesenteries which is regarded as 

 normal for the species, but in Actinia equina only 4"24 per cent, showed 

 variations from the normal arrangement of the mesenteries. In A. eqtcina, 

 therefore, the arrangement of the mesenteries is a much more reliable 

 character for specific diagnosis than it is in Metridium marginatum. 



.Some years ago* 1 suggested the use of the term " plastic " for characters 

 that are variable within the species and " rigid " for characters that are fixed, 

 or show only a small percentage of variations. The term " plastic " suggests 

 that the characters are to some extent moulded or modified in the course 

 of their development by external forces, and that the exact form that they 

 exhibit in the adult must be due in large measure to the environment. 

 There seems to be no doubt that this is true of many of the plastic 

 characters of sedentary animals, although it may be difficult to prove to be 

 true in the case of some of the variable characters of the higher bilaterally 

 * ' Reports of British Association,' 1903, p. 680. 



