The morass shows, in the dry season, the foundations of buildings, regularly arranged, lining streets which intersected at right angles over a considerable part of the Rawa Baleh Gambang. Their disposition has been advanced to support the theory that the population of the Diëng lived in wooden houses, built on those substructures of stone. The theory that the superstructures of stone have been carried away and the submerged substructures left because not so easy to get at, is just as plausible; perhaps a little more so. But whatever they were, temples and priestly or private dwellings of wood or stone, the officiating clergy, their assistants and the inhabitants of the city ministering to their fleshly needs, must have suffered a good deal from the dampness of the soil, the plateau offering already in those early days a field of rich promise for the experiments of hydraulic engineers. Among canals and ditches of less importance, the Guwa Aswotomo, a _cloaca maxima_ some twelve centuries old, still relieves the plain of its superfluous water. According to the legend, for nothing in this locality goes without at least one,--according to the legend then, the subterraneous passage was dug by Aswotomo on his expedition to the Diëng for the purpose of smashing the Pandawas, and nearing Arjuno’s residence he pushed his way up to the surface, from distance to distance, spying how far he had yet to continue his underground march. Descending into one of the peep-holes he made, in a season of extreme drought, I was able to crawl on to the next, through mud and debris which blocked my further progress and, unable to crawl out on a level fifteen or twenty feet lower, the watercourse sloping deeper and deeper down, I had to return to my point of ingress. The glory of this feat diminishes in the light of my knowledge of the circumstance that the Diëng plateau harbours no snakes,[30] save the decorative _nagas_ of temple architecture, and that a companion followed my movements above ground; had we been provided with ropes, we might have carried our work of exploration much further--but that must wait for another time. Of the rare plant which grows nowhere but in Aswotomo’s burrow and owes its growth to his copious perspiration while at his task, a fern possessing rare qualities, highly beneficial to him who pulls it out by the roots, I saw or, rather, felt nothing in groping my way through mire and darkness. Taking its course in a direction inverse to the mole-man’s initial tunnel boring, his Guwa begins at the Arjuno temples as an unpretentious drain and runs, for about half a mile, slanting toward the source of the river Dolok, where Junghuhn has set up two _lingas_. [Illustration: IV. _CHANDI_ BIMO OR WERGODORO ON THE DIËNG PLATEAU (Archaeological Service through Charls and van Es.)]