l6 INFLUENCE OF PARENTAL ALCOHOLISM 



house of three to four rooms. Thus in one Lancashire town 

 out of 2000 random families of all classes 22% had two or 

 less roomed tenements ; 17 % three-roomed houses ; 44 % 

 four-roomed houses ; 17 % five and more roomed houses. 

 But this only shows that the custom is to build four- 

 roomed houses, it does not indicate how many lodgers, 

 or other families even, were crammed into these four- 

 roomed houses. What we do know is that in a separate 

 house town 32% of the operatives and labourers lived in 

 one and two-roomed tenements, 25% in three-roomed 

 tenements and 43% in four or more roomed tenements, 

 but the actual number of rooms used b}^ the family in the 

 latter cases is not known. I cite this result to show 

 that wliile the Edinburgh sample appears at first sight 

 worse than this Enghsh case, the EngHsh case wants far 

 more investigation before it can be said to be better than 

 the Scotch with a different system of housing. In none 

 of the three matters on which stress is laid does Mr. Keynes 

 provide any comparative data, the sole means by which 

 he could have substantiated his assertion that this was 

 a peculiarly differentiated population. 



Now from Glasgow we have details of nearly 73,000 

 children. The schools examined are grouped into four 

 classes. Groups A and B are schools of the poor and 

 poorest districts ; these embrace 50,000 children ; Groups 

 C and D belong to the districts of the better social classes. 

 Group D including four out of the five Higher Grade 

 Schools in the City ; these embrace 23,000 children. Of 

 those 50,000 children no one can say that they represent 

 an exceptionally ' low grade ' population, they are the 

 great bulk of the children of the Glasgow working-class 

 population. And yet 79 % of them come from one- and 

 two-roomed tenements. In fact, of the whole child 



