PREFACE vii 



ground for the statistical meteorologist. On the other hand no statistical investi- 

 gation has been too laborious when it has been undertaken to unravel some 

 physical problem. 



The keynote of my work has been physics and I have attempted to find 

 the physical explanation of each of the meteorological phenomena observed. It 

 will probably be found that many of my physical explanations are not sound 

 and will not bear critical examination, still to me an imperfect physical e^iplana- 

 tion is better than none, for it acts as a thread to bind the facts together and 

 being an object for attack may lead ultimately to a correct explanation while the 

 mere statement of the facts might have been passed unheeded. Those who have 

 the patience to read through this book will find many fascinating unsolved 

 problems and these, I hope, will be incentatives to further investigation both in 

 the study and in the field. 



A word must now be said about the units used in this work. I had to 

 take with me the instruments available and as most of the ordinary meteoro- 

 logical instruments were provided by the London Meteorological Office they were 

 graduated in English units : the thermometers in degrees fahrenheit and the 

 barometers in inches. Thus if I had not used these units in n\v discussion every 

 reading would have had to be converted into another set of units. But what 

 units should these be ? On my return I found the meteorological world vehe- 

 mently discussing the relative values of the fahrenheit, the centigrade and the 

 absolute scales of temperature, and tlie inch, the centimetre and the millibar 

 units of pressure. The controversy still rages and we now have all these units 

 in actual use by official meteorologists. In such an atmosphere of unrest it would 

 have been difficult to justify the immense labour of converting my observations 

 from the units in which they were taken into any other set of units. Therefore 

 in the ordinary meteorological and climatological observations I have used the 

 old English units of fahrenheit degrees in temperature and inches in barometric 

 pressure. On the other hand, in special problems I have used the units which 

 are most frecjuently met with in previous works. My upper air observations have 

 been given in centigrade degrees and metric units, and when I have been 

 discussing the work of continental writers I have followed these units even in ordinary 

 climatological elements. Thus I lay myself open to criticisms of inconsistency, 

 but if my inconsistency helps to facilitate discussion I am quite content. The 

 book has been written in a transitional age and it bears the marks of the transition. 



I have received much help in writing this book, and in every case in which 

 I have asked for data it has been willingly and promptly supplied. I cannot 

 record all the help which I have received in this way but I must acknowledge 

 that which I received from the directors of the meteorological services of Australia 

 and New Zealand. Mr. Hunt in Melbourne and the Rev. D. C. Bates in Wellington 

 extended to nie personal hospitality and placed the resources of their departments 

 at my disposal. The data sent to me by Mr. R. C. Mossman have been of the 

 utmost value and I am grateful to him for the help he has given. Dr. Gilbert 



