1. The nature of the brille or tertiary spectacle in lizards and snakes is discussed.
2. The morphology and relations of the orbital glands, conjunctival or subbrillar space and lachrymal duct system were investigated in ten different species of lizards, two of amphisbaenids, and fifteen of snakes.
3. In all lizards, except one anguimorphine form, Pygopus, not examined by us, and in Sphenodon, the Harderian gland discharges by separate ductules into the conjunctival or sub-brillar space, on the deep surface of the nictitating membrane when this is present. The lachrymal duct always arises by two puncta and canaliculi from the conjunctival space on the deep aspect of the lower lid. This relationship is not influenced by the replacement of the lids by a brille, and it is suggested that this was the condition in the primitive ancestors of the Rynchocephalia and the Lacertilia.
4. In the two amphisbaenids studied, Trogonophis weigmanni and Amphisbaena fuliginosa, a well developed sub-brillar space is present. The lachrymal duct arises from this by a short wide communication, and the Harderian gland discharges both into the sub-brillar space and into the lachrymal duct. With regard to the structures studied, these reptiles exhibit greater ophidian than lacertilian affinities.
5. In snakes varying relationships are found between the Harderian ducts, the sub-brillar space and the lachrymal canaliculus. In the more primitive types (e. g. Constrictor, Cylindrophis) the Harderian gland discharges directly into the sub-brillar space as well as into the lachrymal duct. In “typical” snakes (e. g. colubrines and vipers), the Harderian gland discharges by a single duct into the upper end of the lachrymal duct only, and has no direct communcation with the sub-brillar space. The lachrymal punctum is minute, and the lachrymal canaliculus enters the dilated lachrymal duct in a peculiar and characteristic manner.
6. The lizard, Pygopus, not examined by us, shows conditions which may be regarded as representing a transitional stage between those in typical lizards and in snakes. Here the Harderian gland discharges into the lower of the two lachrymal canaliculi, and not into the sub-brillar space.
7. These findings are discussed in the light of current theories of ophidian phylogeny; the hypothesis of the origin of snakes from burrowing ancestors in particular is considered.