MORPHOLOGICAL CHARACTERS OF TAXONOMIC IMPORTANCE as 



increases as the animal grows in size. As previously mentioned (p. 5) body length cannot be measured 

 in the larger specimens with great accuracy and this no doubt contributes to the considerable scatter of 

 the data plotted in Text-figs. 7 and 8. Even allowing for this scatter, particularly in the data for 

 S. thompsoni, it is obvious from Text-fig. 7 that in the solitary form S. gerlachei has much thinner 

 muscle bands than 5. thompsoni. 



Text-fig. 8, in which data for the aggregate form of each of the two species has been plotted, show 

 that there is again diflFerentiation between S. thompsoni with wide muscle bands and S. gerlachei with 

 thin muscle bands, although the difference is not so marked as it is in the solitary form (see Text- 

 fig. 7). From Text-fig. 8 it is obvious that there is considerable scatter in the observations which 

 makes muscle width in the aggregate form a character of doubtful value for diflferentiating between the 

 two species, even though the data for S. gerlachei consistently fall in the lower part of the graph. 



Table 5. The variation in muscle width in Salpa thompsoni and S. gerlachei expressed as the 

 ratio body length: muscle width, together with the number of observations. 



The results plotted in Text-figs. 7 and 8 may be expressed as numerical ratios of body length and 

 muscle width (Table 5) when it is seen that there are significant diflFerences between the ratios for the 

 solitary forms of the two species. As mentioned above, owing to variability in the data, the ratios for 

 the aggregate form cannot be considered to be reliable. 



The Development of the Stolon in Salpa thompsoni and S. gerlachei 



A consideration of stolon development and budding in the solitary form is outside the scope of this 

 paper and will be described in a later work. There are, however, certain differences between S. thomp- 

 soni and S. gerlachei, both in the size of the chains of buds and the size of the parent solitary individual 

 at the time of release which, with the morphological differences already described, lend support to 

 the view that they are in fact valid species. These differences will be discussed briefly here. 



In the stolon of a mature solitary individual, that is, one in which a chain of aggregates (blastozooids) 

 is about to be released, three zones or blocks of bud formation can be distinguished, the distal two 

 being fully differentiated into a number of segments or buds which can be counted. When the oldest 

 distal block, which I shall refer to as block i , reaches a certain size it is shed in its entirety through a 

 hole in the test so leaving a section of empty tube in the test. In the species to be described the hole 

 formed by the release of a block is sealed off by a tissue ' strand ' of granular appearance which remains 

 as evidence that one block of buds at least has been released. The section of empty tube in the test 

 which was occupied by block i is quickly filled by the growth of block 2 which thus becomes the 

 distal block. It is probable that this block is subsequently released and replaced by block 3 which 

 may in turn on its release be replaced by another block. It will be realized from the foregoing that 

 while it is possible, from the absence of a strand of granular tissue in the test, to deduce with certainty 

 that no blocks have been released it is not possible to say, in the presence of a strand, whether the 



