V'ART.'l.j Mallei: Mud rolcanoe-^s uf Hduin and Ckeduhu. ' 19ri 
mound, so that the latter presents, as before described, the aspect of a heap of 
stony debris on which only the more recently formed and active mud cones 
remain. On the whole, however, the ejections are generally more frequent, or 
more abundant, towards the centre of the mound, which is conseqxiently highest. 
Occasionally, however, the summit is found elsewhere. 
In the case of the second form, the ejections (of sufficiently viscid mud) have 
remained constant to a single vent, around which the entire mass of matter has 
by degrees been built up into one cone of regular outline. In the character of 
the ejected stones, the surrounding casuarina ring, &c., the cones are quite 
similar to the mounds. 
On rare occasions the mud volcanoes are subject to jiaroxysmal eruptions 
Ordinary conditions of erup- of great violence. In ordinary times, however, they 
remain in a state of comparative quiescence. Some¬ 
times action is suspended altogether, but generally a few mud cones, from a few 
inches to as many feet high, are scattered over the mounds, in the majority of 
cases at or near the centre. The craters at the top contain mud in various degrees 
of fluidity. Sometimes it is thin and watery; in other cases thick and viscid. 
The thin mud forms cones of very regxilar outline, but low inclination; sometimes 
not more than 16 degrees: the cones produced from viscid mud, on the contrary, 
rise at steep angles, and are sometimes nearly vertical at the top. In such cases 
they are formed, in part at least, by small quantities of mud spirted out, which 
drop back on the rim and gradually build it up like a wall. Bubbles of 
inflammable gas rise through the mud in the craters, when they are active, in 
greater or loss number. When the mud is very thick the gas issues with a 
peculiar ‘ clucking’ sort of noise, and when sufficiently forcible, produces the 
spirting just alluded to. 
Besides such spirtings, flows of mud issue at times from theci’aters, generally 
finding an exit by breaking down the weakest part of the rim. A stream of some 
magnitude had flowed from the most northern of the Kyauk Phyu conical volca¬ 
noes not long before I visited it. Issuing from a gap in the crater wall, with 
a breadth of 18 inches or 2 feet, it flowed down the side of the cone, with a very 
constant width. At the bottom it got into a small channel, and continued its course 
for a distance of 100 yards from its source. The stream well exhibited that 
canal-like form which sufficiently viscid mud assumes under such circum¬ 
stances. Owing to the sides moving more slowly than the centre, and hence 
having more time to dry, the mud there first becomes too consolidated to flow 
further, and forms banks between which the central portion still flows on. An 
analogous phenomenon is frequent in lava streams. “ A stream of lava,” says 
Mr. Scrope,“ while flowing down any slope, will, owing to its imperfect fluidity, 
usually be thickest towards its centre, and consequently possess a convex cross 
section on its .upper surface, the sides rising as steep banks from the uncovered 
ground adjoining. But when the supply of fresh lava from the vent diminishes 
or entirely ceases (the still liquid interior at the central part of the current 
continuing for some time to flow on, urged by its oum weight above, down any 
slope that offers itself), the upper crust, being unsuppoi-ted, will have a tendency 
to subside in proportion. Hence we often find narrow lava-streams confined 
