explains the general methods adopted for discrimination between species, and needs not to be 

 repeated here. One or two points, however, require comment. 



As in the Biscayan report, a table of (what may be termed) formulae has been furnished 

 for almost every species at different lengths (presumably = ages). In all these tables (i) the 

 first column of figures gives the total length in millimetres: (2) the second, the length of the 

 tail, expressed as a percentage of the total length: (3) the third, the number of jaws (cirrhi, 

 Greifhaken) : (4) the fourth, the number of anterior teeth (or of the only row of teeth in the 

 case of Krohnia): (5) the fifth (when present), the number of posterior teeth. In the case of 

 common species, the tables have been constructed from about 30 — 40 specimens : where the 

 range of variation is considerable at a particular length, it generally implies that several 

 specimens have been under observation. This method, however tedious in the working, appears 

 to the writer a necessary step towards putting a species on a sound footing; not only because 

 the numbers of the armature, and the proportion of tail to body, alter with age, nor merely 

 because the range of individual variation is considerable, but also because it is often difficult 

 to count every tooth, so that the recording of numerous specimens tends to neutralise the 

 errors of observation. As regards the lengths, it also tends to correct the error due to different 

 amounts of contraction. It is very desirable that this method should be adopted during the 

 next few years in the case of every Chaetognath recorded, however common or wellknown, 

 notably in collections from distant seas. 



The characters given below for the identification of species have been deliberately 

 selected with a view to the needs of a naturalist reporting on a collection of Plankton, in 

 which the material often may be immature or imperfectly preserved. They are mainly external 

 and macroscopic, and practically do not take internal anatomy into account. 



Under the heading of 'total length' and 'length of tail', the tail fin has been included 

 in the measurements. The addition of this is not so great as to make comparison with the 

 records of previous observers unreliable, and the actual measurement is made more certain 

 and easy, at a cost of less damage to the specimen by the compasses. 



Drawings of the jaws T ), and teeth when boiled out with caustic alkali from the soft 

 tissues, have been given in many cases, but I doubt whether they have much diagnostic value : 

 especially are the anterior teeth unreliable, for they have so entirely different an appearance 

 according as they are viewed laterally, foreshortened, or in profile. More characteristic is the 

 appearance of the teeth in place, but this view is not always easy to obtain, and is generally 

 impracticable in forms with numerous teeth, such as Sibogae and Bedoti. 



The form of the vestibular ridge seems to prove useful as a diagnostic test of secondary 

 importance ; this does not imply that it is always of identical form, papilla by papilla, even in 

 two specimens of the same species; but that its general character, with high or low, sharp or 

 rounded, numerous or few papillae, etc., etc., is a specific constant. The disadvantage of its 

 use is that it is often difficult to get it into exact profile. 



1) I should have liked to utilise Dr. Krumliach's character of the shape of the tips of the jaws, but — if I may be pardoned 

 for a personal explanation — my eyes are no longer reliable for minute transparent objects under very high powers: the tips have 

 been drawn, however, in some cases when fairly large. 



