igi6. Carpenter — Centipedes and Millipedes.. 167 
the mandibles, as well as a post-maxillary segment behind 
the gnathochilarium. These discoveries show that the 
monoprosthomerous condition of the milHpede head is a 
speciahzed not a primitive condition, and support that 
argreement of head-structure throughout the Arthropoda 
which I advocated in the paper just quoted. 
If the centipedes and millipedes then be relegated to 
distinct classes, the Symphyla must also be regarded as 
a distinct class. These frail little creatures are of high 
interest, as their general appearance suggests that they 
form a connecting link between those primitive insects 
the Thysanura and the centipedes, while in the position of 
the reproductive opening they resemble the millipedes. 
In the segmentation of head and body they seem to agree 
exactly with a thysanurous insect on the one hand and 
with Polyxenus on the other, so that their name — Symphyla 
— well indicates how they form an annectant group with 
affinities at once with millipedes, centipedes and insects. 
A tabular arrangement of the classification advocated in 
this note may be of service to those interested in the subject : 
GLASS DIPLOPODA. 
Order Chiiognatha. 
lulidae. Chordeumidae. Polydasymdae. Glomeridae. 
Order Pauropoda. 
Pauropidae.* 
Order Ps3laphognatha. 
Polyxenidae. 
CLASS SYMPHYLA. 
Scolop endrellid ae . 
CLASS CHILOPODA. 
Order Epimorpha. 
Lithohiidae. Scutlgeridae. * 
Order Anamorpha. 
Scolopendridae. Geophilidae. 
In the above arrangement the series begins with the 
more highly organised of the millipedes, from which a 
* J^ot yet recognised in the Irish fauna. 
