a small increase in number of Class V farms, and small decreases in every other commercial class, plus the large decrease in noncommercial farms mentioned previously.

Beaufort County, South Carolina. --Another Southern geographical area that showed sharp rises in the level-of-living indexes was the coastal plain and tidal areas represented by Beaufort County, South Carolina, where the index rise between 1950 and 1954 was 118 percent. Every item contributed to the increase.

Total farms decreased 27 percent, and percentage of farms reporting tractors rose 14 percent. Economic Classes I, III, and V were larger in 1954, while Class VI and noncommercial farms decreased. As in Breathitt County, part-time farms increased slightly, from 175 to 191. Proportion of tenancy dropped from 12 to 5 percent. Average size of farm increased from 104 to 110 acres. As in Tunica County, diversification of farm products was reflected in a much higher average volume of sales. Acreage in cotton and corn decreased, and acreage in oats, vegetables and cowpeas rose. Number of cattle and volume of milk sold rose sharply.

Perkins County, South Dakota. --Large increases in county indexes also occurred in areas outside the South and the Southern fringe. For instance, in Perkins County, South Dakota, the rise was 45 percent.

This county had a decrease of 8 percent in number of farms, with a rise in average size of farm from 1,808 to 1,946 acres. Farms under 10 acres and farms over 1,000 acres both increased in numbers. A movement upward from Economic Class IV to Class III was evident. Number of farms reporting tractors rose 18 percent, and number reporting trucks, 30 percent. Number of farm operators working off their farm dropped from 241 to 190, and number working off their farm more than 100 days, from 96 to 28. Number of noncommercial farms decreased from 79 to 10, and of them, part-time farmers decreased from 42 to none.

Farmers in this county made a striking shift in type of farming operations between 1950 and 1954. The number of cash-grain farms increased from 167 to 318, while livestock farms decreased from 635 to 452. Sales of milk, butterfat, cattle and all livestock decreased sharply in volume and dollar return, while figures for corn, wheat, oats, rye, flax, and alfalfa increased sharply.

Perkins is the westernmost of a group of relatively low-index counties in northwestern South Dakota which also includes C-Corson and C-Dewey. These combinations had, respectively, 40 and 36 percent increase in level-of-living index between 1950 and 1954. In these combinations the same factors observable in Perkins were present. In this area, apparently, a shift from livestock to mechanized cash-grain farming is helping to raise the index.

Jasper County, Indiana. --This county is an example of a relatively high-index county which had a large index rise between 1950 and 1954. The increase was 26 percent, from 145 to 182, and was sufficient to move Jasper from the second-highest quintile, where it was in the 1945 and 1950 indexes, into the top quintile.

Number of farms decreased 5 percent between the two latest index years, while percentage of farms reporting tractors rose 15 percent, and percentage reporting trucks, 41 percent. Average size of farm rose from 187 to 196 acres, but there were increases in both the very large and very small size-intervals of farms.

As in the South Dakota county previously examined, there was a noticeable shift away from livestock and dairy farming toward field-crop farming, but in Jasper County the shift was more in the nature of a diversification than a drastic shift. And here the increased field-crop farming was mostly in corn and soybeans, while wheat, oats and rye decreased in importance.

Lincoln County, Oregon. --The percentage increase in this county, 44 percent, was one of the largest in the country outside the South. Here increased use of electricity and telephones made large contributions to the index rise.

Lincoln County showed a 21-percent decrease in number of farms. There was a clear-cut trend toward larger, higher-volume, more mechanized farms. Average size of farm rose from 137 to 156 acres, farms reporting tractors rose 51 percent, and there were considerable increases in numbers of farms in the three top economic classes. The nationwide trend toward increased numbers of very large and very small farms was not, however, evident in Lincoln County between 1950 and 1954. Noncommercial farms, parttime farms, and number of operators working off their farms all decreased.

Indexes Drop in a Few Counties

Of the 2, 575 counties and combinations for which 1954 indexes were computed, 106 had decreases from their 1950 indexes. These were distributed in 26 States, and there were such in every geographic division.

Items which are producing effects on the indexes can be readily identified from study of census data. A drop in average value of products sold between 1950 and 1954 goes far toward explaining the decreases in the level-of-living index in some counties, but this is not always the case. Of the 106 counties and combinations of counties which had lowered 1954 indexes, 57 had decreases in average value of products sold, 74 had lower percentages of farms with automobiles, 59 had lower percentages with electricity, and 43 had lower percentages with telephones. Twelve of the counties had decreases only in value of products sold; 20 counties had decreases only in percentage of automobiles; 12 counties, only in percentage of electricity, and 5 counties, only in percentage of telephones. All other counties had decreases on more than one item.

Certain counties and groups of counties which had substantial decreases in level-ofliving indexes are discussed below in order to explore reasons for the index drops.

The dry farming and ranching areas of the West North Central, West South Central, and Mountain Divisions should be treated as a sort of special enclave in discussing counties with lowered indexes between 1950 and 1954. In this area the 1954 indexes show clearly the effects of recent drought and lower farm prices. In these three divisions are found all the counties whose indexes dropped between 1950 and 1954 solely due to decreased average value of products sold, with one exception in east-central California. These were Baca, Colorado; Ford and C-Grant, Kansas; C-Banner, Nebraska; Haskell, Knox, C-Carson, C-Cottle, C-Crockett, C-Dallam, and C-Kimble, Texas. This group shows between 1950 and 1954 larger numbers of very small farms and very large farms, a trend toward more part-time farms and toward a downward rearrangement of the proportions of the economic classes of farms. The average size of farm actually decreased in some of these counties. The proportion of tenancy rose in some of the Texas counties.

In and near the same general area there were a number of counties with considerable decreases in index where average value of products sold dropped and where percentage of farms with automobiles also dropped. Such counties were Towner, North Dakota; CHamilton and C-Morton, Kansas; C-Alamosa, Colorado; C-Broadwater, Montana; Hall, Jack, and C-Archer, Texas; Harrison, West Virginia; and Polk, Florida. All these counties are in dry farming and ranching areas of the plains and Rockies.except for the last two. In the same general area, there were several counties where the same factors were present, but where the index drops were not large.

An examination of 1954 Census of Agriculture data for these counties reveals a pattern of increased numbers of very small farms in all the counties, coupled with an increase in the largest sizes of farms in most of them. The decreases in proportion of tenancy were generally quite small; in Saguache County, Colorado, and in the Montana and Kansas combinations the proportion actually rose. Hall County, Texas, showed a decrease in average size of farm. All except four combined counties--Saguache in Colorado, Meagher in Montana, and Morton and Seward in Kansas--had increases in numbers of part-time farms. There was a general pattern of increased numbers of the lowest economic classes in all these counties and combinations.

In the case of a few widely-scattered counties, there are only lower percentages of farms with telephones to explain the lowered index. This was the case in the following counties: Bureau, Illinois; Ballard, Kentucky; Grafton, New Hampshire; Meigs, Ohio; and Benton, Oregon.

There were other counties where average sales and percentage of farms with telephones dropped, but not percentage of farms with electricity or automobiles. This group of counties included Arapahoe, Colorado; Clinton, Missouri; Mountrail, North Dakota; Beaver, Oklahoma; Brookings, South Dakota; McCulloch and C-Hemphill, Texas. All but Mountrail and Brookings Counties had considerable drops in indexes. Telephones in these counties may have been disconnected when incomes dropped. In most of the counties average value of products sold was very much lower in 1954 than in 1950.

Two adjacent counties in central Florida, Lake and Polk, had index drops between 1950 and 1954 apparently due to a rather unusual influx of new farm operators. Numbers of farms increased from 1, 711 to 2, 920 in Lake and from 3, 256 to 4, 020 in Polk, while the indexes decreased from 129 to 120 and 155 to 148, respectively. The percentage of farms with electricity dropped in Lake County, and percentage of farms reporting automobiles in both counties, as did average value of products. In Lake County, the average size of farms dropped from 146 to 109 acres. There was a large increase in part-time and noncommercial farms, from 545 to 915. In Polk County, also, the average size of farms decreased, from 350 to 310 acres, but the noncommercial farms decreased in number. The county's already disproportionate concentration of farms of from 10 to 29 acres became heavier. This county's farms are mostly fruit-and-nut, and the number of this type increased from 1,374 in 1950 to 2, 503 in 1954. In both Lake and Polk Counties

there was also apparently heavy investment in cattle during this period, with the number of cattle rising by 91 and 41 percent, respectively.

On the edges of the low-income northern Great Lakes area, often referred to as the "cutover, were two combinations of counties with increased average value of sales and yet considerable decreases in index. These were C-Iron, Wisconsin, and C-Crawford, Michigan. The index drops were, respectively, from 126 to 115 and from 124 to 111. The number of farms dropped in both, 14 percent in the Michigan area and 30 percent in the Wisconsin area. The percentage of farms reporting tractors dropped 17 percent in the Michigan combination and 27 percent in the Wisconsin one. Of the counties in the Wisconsin combination, Oneida and Vilas had lower 1954 percentages on all items except average value of products sold; Iron, on all except telephones. The Michigan counties all had increased average values of sales except Roscommon, and increased percentages of telephones, except Crawford, but auto percentages dropped sharply in the counties in this group. The large percentage decreases in electricity in Crawford and Montmorency Counties more than offset small rises in Oscoda and Roscommon. In these two combina – tions, the drop in farms was apparently to a disproportionate extent from among the relatively more prosperous farm operators.

Another area in which some decreased indexes occurred in a few counties was the northeastern dairy belt. Such counties included Chemung and Greene in New York, Washington and the combination C-Caledonia in Vermont, and Strafford in New Hampshire. In this area, the value-of-products item did not contribute much to the index decreases. Decreased percentages of electricity in all but Strafford County, which kept the same percentage, were largely responsible for the drops. The percentages of farms with automobiles also contributed to the lowered indexes in all but Chemung County. The telephone percentage rose slightly in Greene County, was stationary in Chemung, dropped 1 percent in Washington and Strafford, and took a sizeable drop in C-Caledonia.

All these counties lost farms at a rate greater than the national rate of 11 percent between 1950 and 1954. Percentages of farms reporting tractors decreased in the New York counties, decreased sharply in Strafford County, rose by 1 percent in Washington County and by 12 percent in C-Caledonia. Numbers of commercial farms decreased sizeably in all these counties, and in all but Chemung there was an increase in numbers of farms of under 3 acres.

INDEXES ALSO AVAILABLE FOR STATE ECONOMIC AREAS

In addition to indexes for geographic areas, already discussed, indexes for recentlydesignated economic areas within States have also been computed. State economic areas are relatively homogeneous subdivisions of States. They were delineated by the Bureau of the Census in consultation with the former Bureau of Agricultural Economics and other agencies, and have been used in the 1950 and 1954 censuses, in order to provide data for groups of counties with relatively homogeneous economic activity. The areas are groups of counties smaller than States. They are particularly useful for socio-economic comparisons. State economic areas are shown in chart 4.

The indexes for State economic areas point up significant internal differences within States. For instance, indexes range in California from 264 for State economic area 8, which includes the Imperial Valley, to 153 for area 9 in the northern and eastern hinterland. In the wealthy irrigated State economic area 2a of Arizona, where the principal crop is short-staple cotton, the index is 314, highest in the nation, but area 2b in southeastern Arizona has an index of 147. Kentucky area 7 has an index of 145 in contrast with 67 for area 9. State economic area 6 in southern Florida has an index of 178, but areas 1 and 3 in the northwestern corner of the State are similar to the adjoining Deep South, and have indexes of 99 and 95, respectively. In northwestern Illinois, State economic area 1 has an index of 200, but area 11, in the southern tip of the State, shows a relatively low index of 112.

Indexes for State economic areas show a nationwide pattern of higher levels of living in an area near or surrounding a metropolitan center. Some reasons for this pattern are more general availability of electric and telephone service in such areas, proximity to part-time jobs and therefore added income in some cases, and absence of such economic problems as hill-land farming and general isolation. Exceptions to this rule occurred principally in areas where such factors as large volumes of sales, mechanization, irrigation, and patterns of large farms using hired labor are associated with relatively high returns to the individual farm or ranch.

Aroostook County, Maine, comprises that State's economic area 1, and has an index of 193 as against 159 for area 4, which surrounds Portland. However, the latter index is somewhat higher than those for the other two State economic areas of Maine. State economic area 3 in northeastern Colorado has an index of 195, much higher than the one of 153 for area 4, which contains Denver. Area la in the northwestern corner of Iowa and area 8 in southeastern California had higher indexes than any of the areas containing metropolitan centers in their respective States.

The two lowest indexes for State economic areas were in western Alabama, 61 and 66 for areas 6 and 7b, respectively. But Kentucky area 9, in the far eastern hilly section of the State, was almost as low with an index of 67. Other nonmetropolitan State economic areas with indexes from 70 through 79 were West Virginia 4 and Kentucky 8, which adjoin Kentucky 9, South Carolina 8 on the tidal coast, Mississippi 6b, which adjoins the above mentioned Alabama areas, and Mississippi 2, a strip of counties to the east of the Delta.

Outside the South, there were no State economic area indexes for 1954 under 100 (the average county index in 1945) except area 8 in south central Missouri, which adjoins the Arkansas border. Around the fringes of the South are several State economic areas which showed the characteristically low indexes of the Deep South. These appeared in Missouri, Illinois, Indiana, Ohio, and Pennsylvania. Texas, Oklahoma, Maryland, and Florida also showed lower indexes in the areas nearest the southern "core."

Indexes for State economic areas in non-Southern States were below the national average of 140 in several scattered regions. These included State economic area 2 of New Mexico, a desert area with Indian reservations; North Dakota 2a and 2b, part of the general low-index area just east of the Continental Divide which included areas 2b and 3b of Montana and area 1 of South Dakota; Washington 5b and Montana la, a dry and rugged area whose indexes contrasted sharply with the very high indexes of the wheat country of Washington and Oregon just to the west, in the vicinity of Grand Coulee and Bonneville Dams; and, in the "cutover" areas bordering the Great Lakes, Minnesota 2, Wisconsin 1, and Michigan 1, 2, and 4b, where early frosts, isolation, and poor land are associated with relatively low sales volumes and percentages of possession of conveniences.

Three nonmetropolitan State economic areas' indexes dropped between 1950 and 1954. The areas were Colorado 2b, Kansas 1, and Texas lb. The index for Texas area 4 remained the same in 1954 as it had been in 1950. Indexes for these areas are relatively high, being in the highest 40 percent for State economic areas. The decreases in index were not large in any of the areas; there was only a 1 percent decrease in the Colorado area and a 2 percent decrease in the others.

The very high percentage increases in State economic area indexes were concentrated in the South, as might be expected since the area indexes are averages of the county indexes. Notable among these are the increases in Mississippi areas 1 and 2 in the northwestern part of that State, 65 and 73 percent, respectively; Arkansas 8a and 8b, Delta areas with 57 and 62 percent, respectively; areas 6 and 7b in western Alabama with 65 and 53 percent; Florida area 1 in the northwestern corner of the State with 50 percent; areas 8a and 8b in eastern Oklahoma, with 59 and 53 percent; and in northwestern Louisiana, area 1, with 56 percent.