546 Energy Distribution of Diffraction Gratings. 



This phenomenon will have to be taken into account as one 

 of the properties of: an echelette grating with a large, gently 

 sloping surface as one portion of: the element, and mounted 

 in this type of spectrometer, i. e. with fixed collimators. 



Some light may be thrown on the results found by 

 Trowbridge and Wood* with this grating, No. 7. It will 

 be noticed that for wave-length 4'3/a the second order is 

 strong (coming nearly at the angle 23°), while the first and 

 third orders are very weak. The retardation over the 

 element for the first order is 4*3^ ^ or almost exactly one 

 wave-length. The retardation for the third order (situated 

 33° from the central image) may be found by producing the 

 curve beyond the point 23° and w T ill be a like amount. Thus 

 the absence of these two orders agrees with the theory. On 

 the other hand the small amount of energy in the first order 

 for wave-length 8'65 At, coming at the concentration point, 

 is not at all accounted for. However, it will be noticed that 

 even this small amount of energy is all there is for the whole 

 grating (with the exception of the central image), and it is 

 probable that something happened after the observations 

 for No. 6 were taken, but before those of No. 7 were taken, 

 so that the total energy striking the grating was diminished. 



It must be remarked that, since the energy absent from 

 the weak orders is not destroyed, it must be sent into a 

 neighbouring order, and this may either help or hinder the 

 concentration of energy in a particular region in which it 

 may be desired to use the" grating. The case of Trowbridge 

 and Wood's observations on No. 7 for wave-length 4'3 fi is 

 an admirable illustration of the building up of an order at 

 the expense of two neighbouring orders, due to this effect. 

 The retardation over the element at S = 0, the position of the 

 central image, is about two wave-lengths, and this, by reducing 

 the central image, contributes further to the strengthening 

 of the second order. To fully describe the grating No. 7, it 

 would be sufficient to state that it is excellent for work in 

 the region 4/x<\<5//,. 



§7. 

 We have seen that the energy is, in general, thrown out 

 principally at the angle at which it would naturally be 

 reflected, and that there is an effect due to the diffraction of 

 the energy by the single elements. These two properties 

 are not at all independent ; for there can be no energy 

 reflected from a surface in a direction such that the retard- 

 ation over the surface gives rise to a destructive diffraction. 

 * Phil. Mag. [6] vol. xx. pp. 886-898 (1910). 



