56 DR. JOHN SCOULER. voyage I was confirmed in my opinion by an examination of the rocks of Juan Fernandez which have much resem- blance to those of Madeira & where the volcanic relics are more distinct. The soil afforded by the decomposition of volcanic rocks seams to be the most favourable to the growth of the vine, which thrives luxuriantly on the sides of Mount Vesuvius, the hills of Madeira & the valleys of Juan Fernandez. At Madeira the method of cultivating the grape is different from that employed by the French or Italians. A number of poles about 6 feet high are fixed in the earth at short distances & others are put across the tops of the erect ones, by this method [the] vine obtains sufficient support & the air is allowed to circulate freely through the plants. 13th August we left Madeira & on the 18th our prox- imity to the tropic was indicated by the abundance of flying fish we saw. The best way of preserving these fish is to put them into spirits, & this method has the ad- vantage of putting it [in] our power at any future period to examine their internal organization as well as their external characters. In large fishes, however, the expense & inconvenience of this method would be very great; in such cases the more convenient plan is to dry the skin and preserve the viscera in spirits. 23d. The examination and dissection of a turtle we had procured at Madeira, contributed to add a little variety to constant uniformity of our voyage. The sp. of turtle found off Madeira appears to be the T. carucena. It is not to be expected that an animal so well known as the turtle should present anything new to the comparative anato- mist. It is curious that an aquatic animal like the turtle should be furnished with so large a lachrymal gland. This gland is a large granulated organ, loosely connected by cellu[l]ar substance to the internal part of the orbit. The larynx is destitute of epiglottis but the aperture is long and
