2-4 University of California Publications i)i Zoology. [Vol.8 



specimens not well preserved some characters are always dis- 

 torted, so that measurements based upon such material can not 

 be relied upon. I have endeavored to include enough measure- 

 ments in these tables to permit a fairly accurate reconstruction 

 of the outline of each individual considered. All measure- 

 ments, unless otherwise stated, have been made in proportion to 

 total length of specimen as measured from tip of tail-fin to 

 anterior extremity of head. 



In my "Notes on the Identification of the Chatogiihtha" 

 (1908) I stated that length and shape of ovaries, together with 

 size of ova, could not be depended upon for identification 

 because these organs were the last to develop. While this is 

 partly true, it is their distinctive appearance, together with 

 degree of transparency of the species, that catches the eye and 

 affords the first basis of separation. These differences are 

 brought out in plates 1, 2, 5 and 6. On pages 55 to 64, measure- 

 ments of the ovaries are tabulated and discussed. However, 

 owing to immaturity of the individuals of some species, it has 

 been impossible to make these tabulations complete. 



To facilitate identification, all known species of Chaetognatha 

 are here described. Those obtained from the San Diego region 

 are considered at some length, and the remaining species are 

 briefly described at the close of the systematic part of this 

 paper. In addition a few doubtful species of the group, includ- 

 ing Sagitta hispida, S. elcgans, »S\ arcfica, and S. tenuis are 

 redeseribed. 



While most expeditions have scattered their observaticms 

 over a large territory, the ^Marine Biological Station of San 

 Diego has confined its collecting to one locality. Until the pres- 

 ent we have depended upon surface nets of various sizes and 

 meshes, open vertical nets, the Kofoid closing net (see Kofoid, 

 1911), and the Kofoid water bucket (see Kofoid, 1905), for the 

 collection of plankton, and on non-registering centigrade ther- 

 mometers and the Kofoid water bucket for our hydrographic 

 work. While all temperature and salinity determinations made 

 in connection with our hydrographic work are as accurate as 

 the best apparatus permits when in the hands of an expert 

 physicist (see IMcEwen. 1910). our data could not be made com- 



