No abstract

London) said that the natural phenomena with which the Paper was concerned were on a grand scale, but it might be said that as much, if not more, was now known about the Shire and its catchment as about many British rivers. About a year ago he had been surprised to find that it was impossible to obtain any recorded flood information about a large river in the Severn area, and to be told by the people concerned^ that they had never bothered to collect such information.87. The Author had indicated that he did not accept that natural events had an entirely random behaviour. He had attempted, and Mr Morgan felt succeeded, to show that there was a remarkably close relation between the behaviour of sunspot activity, or its incidence, and climatic conditions. It was curious that there should be such an apparently close agreement. 88.If it was possible to predict solar behaviour, then clearly this would enable a short-term forecast to be made of the probable behaviour of rivers. In any event, even if one could forecast only the broad fluctuation of solar activity, to say whether it was increasing or diminishing at any time, a short-term forecast could still be made of probable river activity. 89. He had been in Rhodesia last February when there had been some excitement about the level to which Lake Nyasa would move. In the event, the agreement between the actual and predicted maximum levels had turned out to be much closer than anything he had hoped for at the time.90. The Author's formula for evaporation was a little curious dimensionally! Nevertheless, it was based on extensive data, it seemed to fit the facts, and no e c i a l coefficients were introduced. No doubt more information would be forthcoming in the future to give a further check, although the agreement in Table 4 was remarkably good. Possibly members had been able to apply the formula to information of their own, and it would be interesting to know the results.Sir Thomas Foy (Adviser to British Thompson Houston Co. (Canada) Ltd.) observed that the Author had stressed the delicate nature of the balance of the water account on Lake Nyasa, which was primarily attributable to the high evaporation from the lake relative to the rainfall and inflow. As the Author had stated, the average annual rainfall on the catchment down to Liwonde was about 49 in. The annual evaporation from the lake area was 60-70 in. It would readily be appreciated therefore that the annual water account of the lake area by itself was always in deficit. The free water was the amount contributed by the wet season run-off of the land catchment diminished by this deficit. t DISCUSSION ON LAKE NYASA AND THE RIVER SHIRE 21 3 92. The catchment area, including the lake, was 49,000 sq. miles while the actual area of the lake was 11,500 sq. miles. The ratio of land catchment to lake area was therefore only approximately 3.3 to 1, which was a very small figure. With high and fixed losses from the lake area, it was only to be expected that the effect on the free water in any year would be a magnificati...