NOTES ON THE igi6 ERUPTION OF MAUN A LOA 475 



our route, which indicated the courses of tongues of the new flow. 

 We approached these quite closely as we neared the rift. The 

 trunk of this flow departs from the source at Puu o Keokeo in an 

 east-southeastwardly direction, passing along and around the end 

 of the ridge of ancient pahoehoe that juts out from the cluster of 

 cones; there it swerves to the south-southeastward and spreads 

 down the mountain. 



Near the source the lava was pahoehoe in typical surfaces and 

 in broken crusts and fragments. Except near the source the lava 

 in these branches was a-a wherever they were approached closely 

 enough for this to be determined. At points on the east these 

 tongues were thin, from 5 to 10 or 20 feet deep; but at the 

 south and along the west the lava blocks were piled irregularly 

 from 20 to 40 feet deep, or high, and were still hot and fuming 

 on May 31. 



In passing from the southernmost point to the eastward and 

 northward many thin, narrow tongues (from 5 to 8 or 10 feet in 

 depth, and from 50 to 200 yards in width) were encountered 

 radiating to the southeast and east. These departed from the 

 main stream at higher and higher points. Wherever junctions 

 were seen the departures of these minor branches appeared capri- 

 cious; that is, no evidences of local damming or pooling were seen. 

 Though thinner and much less massive than the more western 

 streams, these were still fuming, and in varying degrees the air 

 above them was in a state of shimmer from heat. However, the 

 emanation of the fumes furnished a more reliable indication of 

 their courses than the heat-disturbed air above them. (Probably 

 tongues, or "toes," project from the main streams on the western 

 side, as others report who viewed them before they ceased flowing; 

 we found no opportunity to follow the margin closely and did not 

 note any conspicuous projections.) 



Any adequate cartographic delineation of the complex out- 

 branching of this Kahuku part of the 19 16 flow can be accomplished 

 only by actual topographic survey. (A reconnaissance survey 

 was made in June, 19 16, by a party under the Hawaii Territory 

 Survey. An adapted, and in some details corrected, modification 

 of this follows, as Fig. 1. A general conception of a long, narrow, 



