30 STATE HORTICULTURAL SOCIETY. 



■corrected only once in ten years, or at the taking of the U. S. census. This 

 without doubt gives the state report an advantage that can not be over- 

 estimated. Michigan is one of the states that is publishing crop reports. 

 The system here must be by this time pretty well understood. Briefly, 

 there is a corps of correspondents throughout the state who report from 

 month to month the condition of growing crops and of livestock, and at 

 the close of the harvest the out-turn of the crops. In addition to this the 

 supervisors make a farm-to-farm canvass each spring at the time of taking 

 the assessment, and report the acres and yield of the crops of the preced- 

 ing year, and the acres in wheat, and the number of each class of livestock 

 ■at the time of taking the assessment. These statistics furnished by 

 supervisors form the basis of the monthly estimates made by crop 

 correspondents. 



This compound system, if it may be so termed, was started as an experi- 

 ment, and has never been deemed to be perfect. The most claimed for it 

 •at its inception was that it was the best that could then be devised. Expe- 

 rience having brought to view some of its defects, it now remains for the 

 people to correct them. What are some of the defects? 



First to be named is the fact that the annual farm statistics are taken by 

 assessing officers. As a class I do not believe more honest or faithful 

 men can be found anywhere than are the supervisors in this state. They 

 are competent and do the best they can. But they are chosen for another 

 and well-defined purpose, viz.: the listing and valuing of property for taxa- 

 tion. That of itself is as much as ought to be required of any man at one 

 time. In this connection it should be noted that when an assessing officer 

 collects statistics of products it quite frequently is the case that the mind 

 of the man assessed somehow connects the statistics with the assessment. 

 to the detriment of the statistics. The assessed fears that a full and 

 honest report of his crops will add to his taxes. 



Again, the assessment is made too early in the spring. It is usually 

 begun immediately after the first Monday in April. At this time spring 

 crops are not planted, and the growing wheat and the fruit buds are not 

 yet in such condition that a report of value can be made respecting them. 



The last defect we will now notice is that the supervisor, being crowded 

 with work, has no time to foot his report. All reports have to be footed 

 in the office of the secretary of state. This throws upon eight or ten 

 ■clerks the work that might be done by about 1,200 men, and of course 

 the publication of the statistics is thereby greatly delayed. Now, as a 

 better system, I would propose: First, that the office of local statistician 

 be appointed in every township and assessment district in the state. He 

 should hold the office so long as he faithfully performed his duties. It 

 should be his duty to collect the farm statistics, and I would add also the 

 statistics of births and the statistics of the defective classes — the insane, 

 deaf and dumb, blind, epileptic, etc., all of which are now collected by the 

 supervisors. He should also be the crop correspondent for his township 

 or district. 



The canvass of the town for the collection of statistics should be made 

 the last week in May and the first week in June. The farm statistics, as 

 soon as the canvass is completed, should be footed by the local statis- 

 tician and forwarded to the secretary of state at Lansing. The statistician 

 should retain a duplicate copy. This is the proposed system, and the 

 whole of it. It could be improved by requiring the statistician, at the 

 time of making the canvass of his township or district, to leave with each 



