144 DIPLOPODA. 
Rogers in Costa Rica and a few from La Palma collected by J. Tristan, because certain differences they 
present from Carl’s description of P. montivagus suggest that they represent a different species, unless his 
diagnosis and figures are inaccurate in certain particulars, as I assume to be the case. 
2. Colour chocolate-brown, with the external half of the keels and the caudal process yellow; antenne and 
legs rather paler brown, the ventral surface and the two basal segments of the legs yellowish-brown. 
Body very gradually attenuated from about the 15th segment forward. Dorsal surface coriaceous, sub- 
granular, the three rows of tubercles only just detectable; back convex; keels of moderate size, and 
inclined downwards and slightly outwards. First tergal plate slightly wider than the head, its lateral 
angle rectangular ; the edge just in front of the angle very weakly tuberculate ; a row of indistinct tubercles 
along the anterior edge of the plate. Keels of 2nd, 3rd, and 4th directed forwards and downwards, nearly 
parallel-sided, the lateral border convex and armed with five small tuberculiform teeth. On the rest 
of the segments the keels have their anterior border lightly convex and the posterior lightly concave, the 
convexity and concavity increasing towards the posterior end of the body; from about the 11th and 
12th, backwards, the anterior border is serrulate, and the posterior border also from about the 7th; 
the lateral border is armed with about five teeth, which are sometimes large and spiniform, sometimes 
smaller and more tuberculiform; and near the middle of the lateral margin there is a shallower or 
deeper angular notch, near the apex of which the pore is situated, its distance from tho notch being 
from once to nearly twice its own diameter according to the depth of the notch; on the keel of the 
19th the pore is near the middle of its upper surface. The upper surface of the keels is more 
granular than the median area of the segments ; the anterior angles of the keels are convex, the posterior 
angles become gradually more and more acute and spiniform posteriorly: those of the 14th very 
slightly surpass the level of the posterior edge of the tergal area; those of the 18th and 19th are 
apically rounded. Caudal process semicircularly rounded, scarcely irregular marginally ; sternal plate 
‘granular, broad, with two widely separated smallish tubercles. Sternal areas of the body granular, 
somewhat deeply sulcate transversely, the sulcus extending right across the middle line; the area in 
front of it lightly sulcate longitudinally and the area behind deeply sulcate vertically, so that the posterior 
sternal area is bicoxiform; sternum of 4th bitubercular, of 5th, 6th, 7th, and 8th quadritubercular, 
the tubercles becoming gradually weaker. 
3. Smaller than the 2, but less convex and with keels better developed. Sternal area of third with a median 
tuberculiform triangular tooth; sternum of sixth excavated to receive the tips of the phallopods, its 
posterior pair of tubercles suppressed. Margin of cavity of phallopods defined behind by a deep groove. 
Phallopods arcuate, crossing apically, the distal half bent inwards and upwards; the auxiliary branch 
curved and apically pointed; the seminal stile much shorter, straight, but pointing in the same direction 
obliquely outwards and downwards. 
Length, 2, from about 70-83 millim., width 11-12. 
” d > 99 ” 60-70 ” ” 8-10. 
Hab. Costa Rica? (Rogers), Carrillo (Underwood), La Palma 1600 metres 3 (Tristan 
and Biolley), Volcan de Turrialba 2000 metres. 
Carl states that the sterna of P. montivagus are unarmed; and Brélemann appears 
to have detected no discrepancy between the specimens he referred to this species and 
those described by Carl. The examples I have seen, however, have well-developed 
spiniform or subspiniform tubercles on some of the anterior sterna. Moreover, neither 
of these authors mentions any peculiarity in the sulcation of the sterna, such as 
meationed above; and, if Carl’s figures be correct, the lateral borders of the keels 
of the 2nd segment are less convex and the anterior angle much more square, and 
the posterior angles of the 11th and 14th considerably more produced and spiniform 
in Carl’s specimens than in those that I have seen. 
In the dentition of the lateral borders of the keels these specimens and P. montivagus 
