Crustacea Decapoda and Stomatopoda. 269 



The specimens agree almost precisely with de Man's account. The only discrep- 

 ancy that I have noted is that the lower border of the rostrum, described as " scarcel}' 

 emarginate at the base," is, as in most species of the genus, distinctly concave 

 above the eye, being at its narrowest only about two-thirds as deep as in the vicinity 

 of the hindmost inferior tooth. 



The rostrum bears from 7 to 10 dorsal teeth, usually 8 or 9 ; the first is remote 

 from the others and is situated on the carapace, the second being as a rule immediately 

 above the hinder hmit of the orbit. On the lower border are from 2 to 5 teeth, nearly 

 always 3. 



The branchiostegal tooth is not very much smaller than the antennal. The 

 outer margin of the basal segment of the antennular peduncle ends in a spine that 

 extends much beyond the produced, setose, antero-external portion of the segment. 

 The two rami of the outer antennular flagellum are fused basally for a distance vary- 

 ing from two-fifths to one half the entire length of the shorter ramus. The antennal 

 scale is about three times as long as wide, narrowed apically and with the distal end 

 of the lamella not extending very far beyond the terminal spine of the outer margin. 



The mandibular palp is composed of only two segments, the joint between 

 the second and third being suppressed. In this respect the species is comparable 

 to the European L. squilla in which precisely the same modification is found. 



The chela of the second peraeopods is nearly always a trifle longer than the 

 carpus, but is occasionally about equal to it, as noted by Ortmann. In reference to 

 the last three pairs of peraeopods de Man remarks, "end of the terminal joint armed 

 with three small and two longer spines." This is doubtless a clerical error, the seg- 

 ment referred to being the propodus. The description, thus amended, applies well 

 enough to the third and fourth pairs ; in the fifth the spinules are much more numer- 

 ous towards the distal end. In the third pair the propodus is one and three quarter 

 times the length of the carpus and two and a third times as long as the dactylus. In 

 the fifth pair the propodus is longer both relatively and actually ; it is about twice 

 the length of the carpus and three and a quarter times as long as the dactylus. The 

 sixth abdominal somite, measured dorsally, is rather more than half the length of the 

 carapace. 



The largest specimen in the collection is a female, 40 mm. in total length. The 

 eggs are of medium size, about 073 x 058 mm. in longer and shorter diameter. 



The rostrum is proportionately longer in small specimens than in adults. 



L. semmelinki, as de Man has remarked, bears a rather close resemblance 

 to Heller's L. modestus; but in the latter species (i) the basal crest of the rostrum is 

 much more elevated, (ii) the interval between the ist and 2nd dorsal teeth of 

 the rostrum is not greater than that between the 2nd and 3rd, (iii) the two rami 

 composing the outer antennular flagellum are fused for a shorter length, (iv) the 

 antennal scale is parallel- sided, not narrowed distally, (v) the mandibular palp is com- 

 posed of three segments, (vi) the fingers of the first peraeopod are nearly one and a 

 half times as long as the palm, (vii) the fingers of the second peraeopod are 

 conspicuously longer than the palm and (viii) the last three peraeopods are 



