J. Appl. Cryst. (1972), 5, 250. [ doi:10.1107/S0021889872099996 ]
The principle of this apparatus is not new (Hope, 1971), but it is reported because of interest shown in its simplicity by visitors to this laboratory. Small samples of the solute are heated in a saturated solution in a poor solvent, and convection and diffusion used to convey the solute to a cooler region where it is deposited as single crystals.