Having been on duty in connection with personnel measures during the past winter, I am presenting herewith a summary of the situation as it now exists.
Two questions had already been under consideration. What measures, if any, were needed to ameliorate the situation occasioned by the “new hump”; that is, the officers taken into the line in 1920 and 1921 from ex-warrant officers and ex-reserve and temporary officers? And what measures could be taken with regard to the older officers in their classes who were beginning to retire from age prior to reaching a position on the list reasonably entitling them to consideration by the Selection Board?
At the same time a bill was introduced in Congress by Mr. Britten, of the Naval Committee (H. R. 10355, December 5, 1924) amending the existing personnel law (Act of August 29, 1916). This bill was referred to the Department and was the occasion of a rather full investigation. It was not approved by the Department, except in a few particulars. At the same time, the result of the study showed that the bill could be modified in some of its details, while preserving its general outlines, and that it would then be workable and include a number of desirable features.
Some of the considerations affecting the question of the “new hump” are the following: These officers occupy positions on the list between the classes that graduated from 1917 to 1920 inclusive. By the Navy Register of 1924 there were about 900 of them in all, of whom about 350 were ex-warrant and ex-commissioned warrant officers. These latter, with long previous service, are much older then their Naval Academy contemporaries, and are already beginning to retire for age. The ex-reserve and temporary officers vary from about the same age as the Naval Academy graduates to ten or fifteen years older. The average would be about five to seven years older.
Under present law these officers are promoted by seniority to the grade of lieutenant commander, thereafter by selection. They are by law immune from age-in-grade retirement until they have served ten years, six years, and eight years, respectively, in the grades of lieutenant commander, commander, and captain. The operation of this requirement will retain these officers for a long period as lieutenant commander, if not selected for promotion, and will largely diminish the number of Naval Academy graduates in the grade who will be qualified for selection by four years’ service in the grade. This situation has caused some uneasiness to the graduates below the “new hump.”
One suggestion has been to repeal the provision of law establishing such periods of service for these officers; but this cannot very well be done without replacing it with something more equitable to all concerned. In the Britten bill, which is based on length of service, it is easy; for the service of the non-graduates is assimilated with the service of the graduates who are their contemporaries on the list. However, if age-in-grade retirement is retained, a feasible course would be to equalize the period of exemption in the three grades by amending existing law to change ten, six and eight to seven, seven, seven; that is, an immunity period of seven years in each of the three grades. This has been recommended by the Department.
It has also been suggested to make all of these officers extra numbers, but it must be borne in mind that there are certain advantages in retaining them on the list as regular numbers as long as they do not interfere unduly with the promotion of graduates now below them. With 900 additional officers, it means nine more rear admirals, thirty-six more captains, sixty-three more commanders, one hundred twenty-six more lieutenant commanders, and six hundred sixty-six more of other grades. Until the Navy reaches its full authorized strength of 5,499, these additional 900 will thus make it possible to keep up materially the strength of all the grades.
A compromise would be to make them extra numbers on reaching the specified age for retirement in any grade, and to retain them thereafter on the active list until they reach the period of ten years’, or six years’, or eight years’ service, according to the grade, and as required by existing law. This would make no change whatever in the prospects of the non-graduates, but would remove them gradually and in time entirely from the path of the graduates. The reduction in the total strength of the line would also be gradual, and would be easily absorbed by the annual graduates of the Naval Academy. It is assumed that enough appointments will be made to the Naval Academy gradually to bring up the line to its authorized strength. This plan would seem to be the simplest, and the most equitable one, for overcoming the apparent difficulties of the “new hump.”
As already stated, the ex-warrant officers are already beginning to retire on reaching the age of sixty-four years, and some of them for thirty years’ service. The number of such retirements will increase with length of time. Also without wishing any of these non-graduates hard luck, they will find it difficult in most cases to pass the professional examinations for promotion; and after reaching the grade of lieutenant commander they will be competing with graduates for selection. A survey of the situation ten or fifteen years hence under existing law, and taking into account all the influences that have been mentioned as tending to smooth out this so called hump, indicates the probability that it will be gradually cleared up without undue delay in promotion to the Naval Academy classes immediately following it. For the rest the situation is not yet acute. The first of the hump will not reach the lieutenant commander’s grade for five or six years, and they will not all be in that grade, where they would hold up the promotion of graduates, for a number of years after that. There is thus time for ameliorative legislation in connection with the “new hump” of non-graduates; but the question should be carefully considered, and any legislation that may be necessary should be enacted while there is yet time.
The other question that was referred to at the outset resolves itself into whether retirement in grades below rear admiral should be for age or length of service. As a matter of fact neither is a sufficient reason for retirement in any of these grades. All the officers in any of the classes are still capable of performing all the duties of the grade. Both age and length of service imply experience, and experience, other things being equal, results in efficiency. The real reason for retirement is that there is a surplus of officers with necessary qualifications for the higher grades. Hence advancement should come first, for as many as are needed, and for those not advanced, some other disposition should be made.
The question, therefore, would seem to be, when should officers be advanced in rank? Should it be age or length of service? This is not considering relative merit, which is still another question. Having decided when the advancement should be made, some go up at such time and the rest go out. The method of determining those to go up or to go out is apart from the present question. It will naturally be some method of selection.
Hence, the real question resolves itself apparently into whether age or length of service should be the criterion for promotion. Put in this way, it would seem that age could not be the criterion, unless all officers entered the service at the same age, and if age could not be the criterion, there is left only length of service.
The average officer has about forty-two years to serve after graduation, if he passes through all the grades. If he serves about seven years in each grade (lumping ensign and junior lieutenant in the first period) it will take him through all the grades in forty-two years. Other periods have been advocated, but this plan seems as reasonable as any, if not more reasonable than any, both for efficiency and for administrative purposes. Hence, if we make promotions at the end of seven-year periods (at the latest) and separate in some way those wanted from those not wanted, the question of age in grade or length of service for retirement disappears entirely. But if we continue to discuss the proper time for retirement in those grades which, as shown above, is aside from the real question, the main arguments that have been advanced seem to be as follows:
Under Age-in-Grade Retirement.—(a) The younger officers in classes feel that they acquired a certain advantage when they entered young, and that they should continue to retain this advantage.
(b) Retirement for age fixes a definite date which is always known and cannot be changed unless the retirement age itself is changed.
(c) In large classes the retirement is spread over several years instead of occurring all at once, which has a double advantage. It eases the strain on administrative details, and it smooths and merges the large classes with those on either side, making greater regularity in the annual retirements.
(d) Owing to changes in the length of the academic course, there have been years with more than one class graduating the same year. This would make a large number of retirements at the same time under length-of-service retirement.
(e) If officers are passed over in a lower grade, some of them under age-in-grade retirement may still get their promotion eventually. If the change is made to a length-of-service requirement, then all of them are apt to go out before they reach a point at which they might be considered for further promotion. .
(f) Change from age-in-grade to length-of-service retirement does not improve promotion on the whole; there will be the same average percentage to go out in each grade under either system.
(g) Change to length of service will benefit temporarily a few officers who are old for their classes, but at the expense of others who are younger.
(h) It is not necessary to change to length-of-service to give even the older officers a chance to be considered by the Selection Board. This may be done by changing the percentages in the grades and by spacing the service in the grades more evenly.
Under Length-of-Service Retirement.—This question has arisen owing to the wide variation in age in the same Naval Academy classes. There was possibly a reason originally for this variation. The course at the Academy was four years, and each congressman had only one appointment. Hence normally, not considering failures, there would be an oppointment every four years. To avoid disqualifying any of the boys in the district, the limit of age would have to be the same; that is, four years. This reason seems to have been lost sight of since the number of appointments has been materially increased. There is no reason now why the age limits should not be reduced at once to two years, say seventeen to nineteen. If this were done there would not be any very great difference between retirement for age, or for length of service.
But to go back to the existing navy list. An assumption has been made that a variation of as much as four years is permissible in the same class, presumably without impairment of efficiency. This holds at the Naval Academy, and through the grades of ensign, lieutenant (j.g.) and lieutenant, where promotion is by seniority. In the grade of lieutenant commander, the assumption is changed, and the older members of classes must now be assumed to be less efficient on the whole, and must retire earlier than the younger members. That is, the original assumption is at this point discarded.
Of course, others things being equal, the normal assumption is that experience goes with service, and more experience means more efficiency. Length of service then, not age, would be the normal criterion for promotion, those who are in excess and not promoted after a stated length of service to be retired.
Taking up the sub-paragraphs under age-in-grade retirement:
(a) The young officers in classes did certainly acquire this advantage. Under seniority promotion they reached each grade younger, and served longer as flag officers. This result was inherent to the system and was accepted as such. It was not in itself to the disadvantage of any older officer in the same class, who went up in regular order, in accordance with his standing in the class. But under the system of promotion by selection and age-in-grade retirement, this advantage was much accentuated and was gained to the corresponding disadvantages of the older officers. The real criterion is efficiency. Efficiency, as has been seen, other things being equal, goes with experience, and experience goes with length of service. This points to promotion for length of service and retirement for those not promoted. It may be remarked that the younger officers, even with the change to a length-of-service requirement, still have the advantage of longer service as flag officers.
(b) This is a valid advantage to age-in-grade retirement. However, in length-of-service retirement the difficulties may be overcome by assuming that each class graduated four years after the date of entry. This puts all classes on an equal footing, whatever the date of actual graduation. Once this date is fixed it is permanent and does not vary.
(c) This is valid, but it is an advantage that is gained entirely at the expense of the older officers in each class and to the benefit of the younger officers, as it gives them more opportunities for consideration by the Selection Board. As for administrative objections in retiring all of a class at once, this is not serious. Those retired are the ones not promoted. They are in excess of the actual needs. They are retired because there is no place for them. As for reduction in total strength, it occurs at the same time the new class of ensigns is commissioned, which at once fills up these and other vacancies.
(d) As has been explained, classes in the past have graduated in from three to six years, and sometimes there have been two classes to graduate in the same year. But only one class has entered each year, and by considering that all classes have graduated four years later than the date of entry, this difficulty is overcome. This is provided for in the bills based on length of service.
(e) It is of course true that if an officer is passed over by the Selection Board he loses his relative place on the list. This is true for either method of retirement. In age-in-grade retirement the younger men may get another chance, but it is at the expense of the older men; another case where the older men in the class are put at a disadvantage through no fault of their own. The average is of course the same in either method.
(f) This is true, but no claim is made that change to length-of-service retirement will improve promotion. This can only be accomplished by better spacing and change in the percentages in the grades. Even if this were done, say with an even spacing of seven years in all the grades (ensigns and junior lieutenants seven years combined), then under age-in-grade retirement, in the grades from which selections are made, with four years required for qualification, officers would have on the average three chances for consideration by the Selection Board. But as the age varies in the classes by four years, the men of average age would have three chances, the youngest men five chances, the oldest men one chance. Under length-of-service retirement all would have three chances. The object of any change from age-in-grade retirement to length-of-service retirement is to place the young and old men in the same class on an equality, which is what was done when they entered the Naval Academy, that is to give them all an equal opportunity.
(g) This could be put differently. The effect will be to restore to the older men in the classes the equal oportunity, which they originally had and which they do not now enjoy.
(h) This has already been answered in paragraph (f) above. Proper spacing would give all officers at least one chance, but to do this the younger officers would have five chances, thus retaining their present advantages.
The general principles applicable to promotion may now be taken up. Figure 1 shows the problem in its simplest form. This is a diagram in which the successive graduating classes are plotted as ordinates at intervals of a year. Each class suffers a certain annual attrition, as it has come to be called. This means the annual percentage loss from all causes except age retirement (final, or age-in-grade). It includes deaths, resignations, dismissals, and disability or voluntary retirements. It used to be about three per cent annually in all parts of the list prior to the World War. There were more resignations early, more deaths and disability and voluntary retirements later, making the percentage in the successive classes practically uniform.
Since the World War it has run up to a total of about five per cent due largely to early resignations. This indicates some unrest or uneasiness among the younger officers, which will probably grow less with time, and especially if the promotion laws are so modified as to afford an equal opportunity to all, really the American principle in all walks of life. The diagram, therefore, goes back to the old normal attrition of three per cent; that is, each class loses three per cent of itself each year. This results in the upper limiting curve of the diagram. (Methods of solving all the problems arising from these assumptions are given in an article by the writer in No. 260, October, 1924, of the Institute Proceedings.)
If promotion is by seniority alone, and the percentage in the grades are as in existing law, then areas may be marked off to correspond; that is, each area equal to the sum of the ordinates at unit intervals. The result shows the ages at which promotion could be expected, assuming twenty-two as the age at graduation; 33.7 to lieutenant, 47.1 to lieutenant commander, 55.1 to commander, 60 to captain, 63.2 to rear admiral, and retirement at sixty-four, clearly an impossible situation.
So something must be done to get officers into the higher grades in time to familiarize themselves with the duties of those grades, and to acquire the experience in each lower grade needed to fit them for the duties of the next higher grade. The total service is about forty-two years (twenty-two to sixty-four). It is beginning to be recognized that the most logical spacing, both for administrative purposes and for training and acquiring the experience for the next higher grade, will be equal periods in all the grades. Ensigns serve by law three years, but for spacing they may be grouped with lieutenants, junior grade. As there are five other grades, this makes six spaces, or seven years in each.
Fig. 2 shows what would happen with such spacing, and with percentages as in existing law. Of course the total is no longer the same as in Fig. 1. Either the entries must be cut down for Fig. 1, or increased for Fig. 2, to preserve the same total. The area between the attrition curve and the irregular line marking the upper limit of the different grades, as shown in Fig. 2, represents the surplus of officers of ages suitable for the duties of the higher grades. This surplus will have been disposed of in some way, in successive steps. The relative difference between the number reaching the top of each grade and the number required for entering the next higher grade shows the number that has been eliminated at that point.
How to eliminate this surplus, and what to do with them, are and have been the main subject of discussion in all proposed plans. These questions will not be taken up here, except incidentally. (The method of elimination was discussed rather fully by the present writer in No. 258, August, 1924 of the Institute Proceedings.)
Captain E. H. Campbell, in his article in No. 264, February, 1925, Proceedings, was probably the first to suggest the even spacing. It is so logical that it appeals at once to those considering the various problems. The wonder is that it has not been proposed earlier. Also, he disposes of the surplus by transferring them to the Fleet Naval Reserve. This is a happy solution, both for the officers concerned, and for presentation to Congress. Our legislators do not look with favor on transferring able bodied officers to the retired list. In fact, in any further legislation affecting the Naval Reserves, it might be advantageous to simplify the transfer of officers to the Fleet Naval Reserve, voluntarily or otherwise. At present, to enroll in the Reserve, they must first resign from the Navy.
Fig. 3 is the actual navy list compared with the normal working of the present law. The percentages are as in Figs. 1 and 2, but the spacing is different. The irregular line shows the strength of the various classes (from the Navy Register of 1924) and down to the class of 1924, totaling about 4,842. The class notation here used is four years later than the date of entry. The reason for this is that the Naval Academy course has varied by law in the past between six and three years. Hence, by taking a date four years later than the date of entry, it puts them all on the same footing. The broken line shows the total of the non-graduates.
The diagram showing the normal working of the present law supposes equal entries each year, an attrition of three per cent (the old normal rate), and a total equal to the total of the present list. Since the two areas are equal, it follows that the part of the actual list above the normal attrition line, plus the non-graduates, will be just sufficient to fill up the shortage in the middle and upper part of the list. Promotion from the grade of ensign is by length of service (three years), thereafter by seniority to the grade of lieutenant commander, and thereafter by selection and age-in-grade retirement for those not selected. The spacing is the direct result of seniority first, and thereafter of the ages for age-in-grade retirement. The amount of drop in the successive steps of the diagram shows the average number of age-in-grade retirements to be expected from each grade annually in a total list equal to that shown, about 4,842. (The full authorized strength is 5,499. The spacing would not be changed, but the ordinates would be proportionately increased. For method of calculating, see No. 260 of the Proceedings.)
Promotion at present is much in advance of that that will be reached later, this owing to the relatively small size of the classes from 1890 to 1917 inclusive. The result in the time of reaching the grades from lieutenant to captain is as follows:
Lieutenant Now 6 years When normal 9 years
Lieutenant Commander Now 10 years When normal 18 years
Commander Now 16 years When normal 23 years
Captain Now 22 years When normal 28 years
The rate of promotion when normal will be thus much too slow. Another very great disadvantage is that with only five years’ service in the grades of lieutenant commander and commander, and a qualification service of four years for selection, there will be left only one year on the average for each class to be available for selection. This means that, with a variation of four years in age in each class, the average men will have one year, the youngest will have three years, while the oldest will be retired for age when only three years in the grade, and lacking one year of the four years’ service for qualification. This is a serious situation for the older men, who do not thus have the equal opportunity to which they might be considered entitled. This applies less to commanders than to lieutenant commanders, for most of the older men in the classes will have been already eliminated in the lower grade at the age of forty-five.
Of the plans submitted about this time is one by Captain Taussig, December 1, 1924, shown in Fig. 4. This is practically the same as his plan in No. 254, April, 1924, Proceedings. It is based on present law, but modifies slightly the percentages and the ages for retirement in the different grades. It gives a better spacing and also meets the need for more commanders and more lieutenant commanders. The excess in the grades of lieutenant and below shown by the drop in these grades, is removed by selection out. These officers are not retired, but are honorably discharged with a lump sum gratuity, varying with the rank. The spacing in the grades of lieutenant commander, commander and captain insures that each officer shall have one or more chances to be considered by the Selection Board, a very desirable modification. The plan is a considerable improvement on the present law and is interesting as showing the drift of service opinion as to the necessity of a change.
The Britten Bill as introduced, H. R. 10355, of December 5, 1924, is shown in Fig. 5. The attrition is taken as four per cent, which is a mean between the actual present percentage and the old normal percentage. The amount of attrition allowed in the calculation may vary somewhat without affecting the spacing. If there is more attrition, there will be less forced retirement, and vice versa.
This bill is based on promotion by selection prior to a stated length of service for each grade, counting from date of graduation, and retirement of those not so selected. The percentages in the grades are as in the present law. Selection is extended down two grades. Lieutenants and lieutenants (j.g.) not selected are not retired, but instead, are given a lump sum gratuity varying with length of service.
The advantage of these diagrams is that they reveal the main features at a glance. First, we notice that the forced retirements (shown by the drop in each grade) are quite uneven. From lieutenant commander to commander there is, in fact, no drop, but a slight rise. Each grade ought to be a reservoir for the grade above. Theoretically, the reduction should be uniform throughout all the grades. This avoids taxing any of them unduly, and retains their reserve strength unimpaired, and available in case of emergency.
Next we notice that the spacing, while a great improvement on the normal existing law, Fig. 3, still has the defect of only five years in the grade of commander, allowing only one year of availability for consideration by the Selection Board. The two grades below, however, have a spacing of eight years each, which is longer than necessary. By taking away a year from each of these, and giving it to the commander’s grade, there would result the ideal spacing of seven years throughout the whole list.
The new features of the bill are thus retirement in certain grades, if not selected, after a stated length of service since graduation, instead of for age, and the extension of selection to lieutenants and lieutenants (j.g.).
The Department did not approve of substituting length of service for age in grade as a basis for retirement, at least at this time, for the reason that the present law had not been in operation long enough to demonstrate in practice that it is or is not satisfactory.
The Department approved of extending selection to lieutenants, but not to lieutenants (j.g.). In another year there will be no excess below the grade of lieutenant, hence there will be no object in eliminating lieutenants (j.g.), other than those who are found to be not qualified, upon examination, as required by law.
The Department’s bill recommended that the forced retirements in any grade should not exceed ten per cent of the grade in any one year, the excess, if any, of those who would be otherwise retired, to be retained for consideration by the next following Selection Board.
Sec. 2 changed the date of expiration of four years’ service in the grade necessary for consideration by the Selection Board from November 30 to July 1, as more in keeping with the present practice of convening the Selection Board in June.
Sec. 3 changed the proviso of the Act of June 4, 1920, in regard to age-in-grade retirements of officers appointed from sources other than the Naval Academy, exempting them from such retirement until they should have served ten years as a lieutenant commander, six years as a commander, or eight years as a captain,to a uniform requirement of seven years in each of these three grades. Ex-warrant officers not promoted were to have the option of reverting to their warrant status.
Sec. 4 made lieutenants ineligible for promotion after fourteen years of service since graduation, and non-graduates counted their service with the graduate next below them on the list. Lieutenants not selected for promotion were wholly retired with two years’ pay. Service since graduation was to count from July i of the year the class would have graduated if it had completed a four years’ course.
All of the above sections were features of the modified Britten plan, Fig. 6, now to be described. The Britten Bill, Fig. 5, was referred by the Department to the Bureau of Navigation and the General Board for comment and recommendation. The Bureau of Navigation recommended a number of changes in the details of the plan to improve its working, without change in the principles on which the bill was based. The General Board commented on the proposed changes, but did not recommend the passage of the bill, for the same reason set forth in the Department’s action, that the existing law had not been in operation long enough to decide whether or not it were satisfactory. The recommendations and comments resulted in the modified Britten plan.
As will be seen from Fig. 6, there is in the modified plan an even spacing of seven years in all the grades (ensigns and lieutenants (j.g.), grouped in the first period). Also the drop (forced retirements) in the different grades is reasonably uniform; that is about the same proportion of the class at the top of each grade. The drop from captain to rear admiral is necessarily more, for there are four captains to one rear admiral.
In order to secure this uniform drop, the commanders have been mcreased one per cent and the lieutenant commanders one per Cent, reducing the three lower grades combined by a corresponding amount. A still better percentage distribution would be that shown by the broken line and the figures in parentheses, a further 'ncrease of one per cent in the lieutenant commanders, and a further decrease of one per cent in the three lower grades.
The percentages in the different plans are compared in the following tabulation.
The parenthesis figures are preferred, and are about a mean of the other two plans. As to the increase in the grades of commander and lieutenant commander, they are very slight, relatively; and by the testimony of all the competent authorities, they are needed, or more than needed, for the actual duties of these grades.
As the Britten plan is based on promotion prior to a certain length of service for each grade, and retirement of those not selected, it is easy to arrange for the non-graduates, who count their service with the graduate next below them. As in some years there would be a large number of retirements, both of graduates and non-graduates, it has been thought desirable to spread these retirements over several years by a proviso in the modified plan like the one in the Navy Department bill, limiting such retirements to ten per cent of the grade in any one year. All of the special features in that bill, as previously enumerated, are also incorporated in the modified Britten plan.
The modified plan has also an additional feature, which it is believed will be of much benefit. In the lower grades, before selection is reached, there are always a few who are not up to the mark, and who manage in one way or another to get past the examining board when they come up for lieutenant (j.g.), and lieutenant. The board now has no option except to pass them or to drop them with not more than one year’s pay. This does well enough, perhaps, for ensigns, but for lieutenants (j.g.), the modified plan gives them eighteen months’ pay, and for lieutenants and lieutenant commanders two years’ pay. In all of these grades they are not dropped hereafter, but are wholly retired. This increases materially the alternatives available to the examining board, and will no doubt help to dispose of cases of the above nature, who would otherwise be left to drift along into the higher grades. The Department’s bill, Sec. 4, provides similarly for lieutenants not selected for promotion, who are wholly retired with two years’ pay.
The modified Britten plan, as above described, and as shown in Fig. 6 (in which the broken line contour is preferred), is probably as satisfactory as any plan that can be devised based on the principles of the present law. There remains a single feature of unequal opportunity that cannot be entirely corrected under those principles. The present Naval Academy classes are quite uneven in size (See Fig. 3). As those classes arrive at the head of the various grades they must be promoted or retired. The vacancies above them tend to become more and more uniform, especially in the higher grades. Thus the proportion of retirements in the different classes will vary very materially. Result, unequal opportunity. This is softened considerably in the modified Britten plan by providing three full classes at the top of each grade from which selections may be made, allowing them to be made proportionately to the size of the classes if so desired, and by the ten per cent proviso, which spreads unusually large retirements over several years. This is much better than the present law, but it does not remove entirely the defect of unequal opportunity.
There is a possible solution. When a large class reaches the top of any grade, and the promotion of all of them would cause an excess in the grade above; then, instead of reducing such excess by retirements from this one class alone, reduce it by retirements from the whole grade above. This is not possible except by combining some form of selection out, with the present selection up. The simplest form of selection out is by reverse effect; as for instance, after selection up, continue the selection in each grade to constitute an eligibility list for future selection and to include as many as are allowed to fill up the grade, all not on the eligible list to retire. Thus retirements could be made throughout the whole grade, instead of entirely from the class at the top of the grade below. This would remove the last remaining vestige of unequal opportunity, as nearly as it can be accomplished by provisions of law, and would result in a promotion plan that would be as equitable as could be made. There might be further changes, but they would be in the details, such as percentages, length of service in the grades, proportion of selection up to selection out (elimination by non-inclusion in the eligibility list). There would be no necessity for further change in the principles. Not all officers can expect to be promoted to the higher grades, but they can reasonably expect that in the selections for such promotions they will enjoy an equal opportunity with their contemporaries in the same part of the list.