metropolitan areas of 300,000 or more. For example, federal court has been held only intermittently in Lansing, Michigan in recent years. Lansing is the state capital, it is seventy miles from Grand Rapids (the nearest court location), and the metropolitan area in 1970 had a population of just under 400,000 people. A similar candidate on a similar basis would be Bakersfield,

California, nearly one hundred miles from Fresno, with a metropolitan area population of 329,000. Bakersfield currently is not a statutory court location.

-As suggested above, places more than one hundred miles from the nearest court location, with populations of more than 100,000, are at least worth consideration as new locations.

Sample Schedule of Court Terms in the Northern District of Alabama (Civil Only: All Criminal Matters Heard in Birmingham)

We are aware of no district that has found an entirely equitable way to divide the case load assigned to the various divisions-except the two districts we visited in which all judges live in the same city. In those two districts the problem is simple: in New Mexico, all cases are assigned randomly, and in Northern Alabama, judges have rotating assignments to the outlying divisions. In Northern Alabama, a judge assigned to a less burdensome division will eventually find his way to the more burdensome

ones.

Southern Florida uses a rough application of the case weight system in an attempt to equalize the distribution of difficult cases. Patent, trademark, antitrust, securities, and school desegregation cases are separately assigned to judges at random, without regard to the division of origin. Other cases are assigned to the judge responsible for a particular location, with one remarkable exception to be discussed below. While this system may equalize the distribution of cases in certain highly visible case types, of course it does nothing for the majority of cases. Substantial inequity among the judges may well appear, in spite of this assignment system, and this possibility is a likely source of contention in any district.

Theoretically, it would be possible to establish a system that equalizes the case load assigned to each judge, based on weighted filings. This could work approximately the way Southern Florida handles assignments to the judge in West Palm Beach. At the end of each year, the court determines whether his assignments were equal to the others', and in the new year,

an adjustment is made based on any discrepancy between his total case load and that of Miami judges. Of course, this has twin disadvantages: the adjustment is made well after the imbalance on which it is based, and it is only as equitable as the assumed equivalence of each West Palm Beach filing to each Miami filing. Using case weights for this calculation would be complex and time-consuming for the clerk's office, and would partially remedy the second fault but not the first. It would refine the determination of case equivalence to the degree that the case weights reflect differences in average burden in the two cities involved. Since the case weights were not designed for this purpose, probably little could be accomplished that would justify the time and expense involved in carrying out the necessary calculations.

There is one exception to Southern Florida's practice of assigning cases to a judge who is responsible for a particular division. Fort Lauderdale cases are assigned at random to Fort Lauderdale and Miami judges, without special preference to the Fort Lauderdale courthouse, judge, or clerk's office. Since there are more Miami cases than Fort Lauderdale cases, a Fort Lauderdale case is likely to be assigned to a Miami judge. Since the distance between the two is less than thirty miles, there is no great inconvenience involved. The fact that this assignment is even conceivable, however, suggests the obvious: there is no evident justification for a separate facility in Fort Lauderdale, Florida.