THE WAYS OF LIFE 221 



were sometimes made before they got right again. The 

 sortie usually began about sun-down (6 p.m.), but earlier 

 if it was a dull afternoon ; there seemed always to be 

 hesitation and caution at first ; a number of soldiers acted 

 as scouts, discovering the best tree ; and there was always 

 that turning back of certain individuals who kept the 

 main body in touch with the advance guard. The orders 

 seemed to be given through the antennse or by a quivering 

 of the whole body. The retreat usually began at dawn 

 and lasted for four or five hours. Bscherich notes that 

 most of the return journeys ended about nine or ten o'clock 

 in the morning. Photographs of the sortie (taken by 

 magnesium flashhght) and of the retreat (taken in daylight) 

 showed that the long troop of workers marched between 

 two lines of soldiers who kept their heads turned outwards. 

 As to numbers, Escherich computed that a vigorous 

 band, crowding past at the rate of about 600 in a minute, 

 would comprise about 200,000 individuals. Professor 

 Bugnion counted about a thousand to a yard, and as the 

 army took five hours to file past at the rate of a yard per 

 minute, there must have been about 300,000 individuals. 

 There were over two hundred soldiers to every thousand 

 workers. Professor Escherich has shown that the number 

 of soldiers guarding a march varies greatly with the danger. 

 When the risks are great the soldiers stand within an 

 antenna-length of one another so that they are always in 

 touch. One morning the returning troop was harassed 

 by the little true ant previously mentioned. Professor 

 Bugnion counted two hundred soldiers on a length of four 

 feet forming at a critical point a living wall covering the 

 retreat of the black workers. It may be noted that 

 the species here dealt with does not eat wood, but subsists 



