Professor T. Zammit was so kind as to invite me, during a short visit to Malta in March and April 1921, to conduct supplementary excavations under the torba floors of the sanctuary at Hal-Tarxien, which he had discovered and excavated. The results are not without interest both for the history of the building and from the nature of the objects found. They bear out Professor Zammit's conclusions as to the relative date of the various portions of the building; and we may add that the spiral decorations and small niches found in the temple of the second period all appear to belong to the latest (third) period in the history of the whole. It also unfortunately seems clear that we have not, as I had hoped, acquired any information to help us in the dating of the various forms and decorations which we find in the pottery of Malta. The excavations in those parts of the building which belonged to the first and second periods revealed in almost every case the existence of an earlier floor below that which had previously been cleared. Taking the earliest building first, we found that the slabs in the right-hand apses BB, DD (which are alone preserved, the left-hand apses having been destroyed by subsequent alterations) rested upon the rock, which had been cut away so as to follow their curve, and were kept in place by inclination against one another, smaller stones being placed to block up the interstices between them.