dissenting:
This is nothing short of an attempt by the City of Aberdeen to sneak in under the wire and conduct a liquor referendum using 1980 census figures. The city knew full well that 1990 preliminary population results indicated that the city had insufficient population under the statute to even conduct a liquor referendum. Nevertheless, the city chose to circumvent the law and hold a liquor referendum prior to the “officially released” new census data. In this first impression issue, the majority adopts the minority view of our sister states with which I cannot agree and therefore, must dissent.
Mfiss.Code Ann. § 67-1-14 (Supp.1990) required at least a population of 7,000 for cities to hold a local option election. The 1980 census showed Aberdeen to have a population of 7,184 residents. However, the preliminary figures made public from the 1990 census showed a population of only 6,771. The preliminary 1990 figures had been released by the Bureau of Census, properly and legally published, all in accordance with the “latest federal census.” Aberdeen’s population decline, as shown by the new 1990 census, were of obvious public notoriety due to the four publications in the local newspaper. In a hurried attempt to get around the new census figures, the city fathers claimed the old figures under the 1980 census were controlling.
We must first answer the question of what is the Legislature’s intent as to which census *212should control, the 1980 or 1990 census. The use of the word “latest” in the Mississippi Code Ann. § 67-1-14 (Supp.1990), indicates that the 1990 census figures, even though preliminary, should control. The 1990 census should also be controlling under Section 67-1-14, enacted after the 1990 decennial census date. In City of Detroit v. Nims, 330 Mich. 239, 47 N.W.2d 4, 9 (1961), the court held that legislative intent, when ascertainable, is the guiding star. That court, addressing the use of the word “latest census” in the Michigan statute, also noted that the use of the word latest “disclosed a clear intent ... distributions should be made on a basis as nearly in accord with actual population as possible.” Id., 47 N.W.2d at 9. The Michigan court thus gave immediate effect to a preliminary report (released August 1, 1950) of the 1950 census results, rather than wait some eight more months until the “official” results were filed.
In Forde v. Owens, 160 S.C. 168, 158 S.E. 147, 148 (1931), according to statute, the compensation of the mayor and councilmen of the City of Columbia could double whenever it was determined by the “last preceding published United States Census” that the city’s population had attained 50,001. The preliminary publication announced by the Board of Census were 50,195. The city officials certified that the city had exceeded the numerical amount of population required by the South Carolina statute and began drawing salaries at twice the rate previously received. The city officials were promptly challenged. The South Carolina Supreme Court, in holding that the governing officials did not act too soon, noted that the census was “published” when the results were made known “by bulletins published in the newspapers” ... and that it was “not necessary to wait” until there have been compiled and bound certain volumes constituting the final official report of the Bureau ... to the Secretary of Commerce. Id., 158 S.E. at 150.
In Board of Com’rs of Lorain County v. City of Elyria, 174 Ohio St. 135, 21 Ohio Ops.2d 393, 187 N.E.2d 33 (1962), where a village and a city joined in petitioning to annex part of the village to the city, the Secretary of State proclaimed that the village had already grown in excess of 5,000 residents according to the latest 1960 census, and was therefore, now a city, not a village. Id., 187 N.E.2d at 33-34. In holding that this information met Ohio’s statutory language requiring the Secretary of State to act when the census count was “officially made known to the Secretary of State,” the court found that this informal certification by the Director of the Bureau of Census was sufficient. Id. at 35. The court noted that the Secretary of State did not have to wait “until the final official declaration of the census [w]as sent to him.” Id. Newspaper legal publications are informal channels, but the focus is the authority of the source, and the source is the Bureau of Census. If preliminary released figures were considered “official” in City of Elyria, they most definitely should be considered the “latest ” census under Miss.Code Ann. § 67-1-14.
In State ex rel. Retonaz v. Mitchell, 115 S.W. 1098 (Mo.App.1908), the Missouri Court of Appeals considered the eligibility of a city, based upon population, to opt for a liquor law that may or may not have been identical or different from that chosen by the county. The county held an election on March 7, 1908, which was held improper because a local census completed February 7, 1908, showed that the city had a population sufficient enough to qualify under the same statute to conduct its own city election on enacting a local option law. Id. at 1100.
This issue is not confined to this century. The case of Nelson v. Edwards, 55 Tex. 389 (1881), as in Aberdeen, involved a county that was losing population. The county had a public office of tax collector and elected a tax collector during November 1880 elections. The census of 1880 showed a decline in population sufficient to drop the county to a class where the sheriff was to be both sheriff and tax collector. The final census figures had not been released and thus the dispute arose. The court, in finding that the enumerator’s preliminary list filed in the office of the county court, was sufficient evidence of the 1880 population, held that the 1880 census governed as to whether the county should elect a tax collector or whether such position was no longer necessary. 55 Tex. at 393. *213The court invalidated the November 1880 election.
State ex rel. Clark v. Board of Education of City of Salina, 148 Kan. 632, 84 P.2d 507 (1938), is also analogous to the case at bar. A city in Kansas attempted to beat the new census by conducting a vote of a school bond the month prior to the new census being officially published. The court followed the principle of practicality rather than formality and held that the most recent census should guide the city’s decision despite the fact that it had not been published officially. The most recent census was the most recently taken census. Id., 84 P.2d at 508-09.
In City of Rochester v. Monroe County, 93 A.D.2d 625, 628, 462 N.Y.S.2d 939, 941 (N.Y.App.Div.1983), the New York appeals court noted no “statutory requirement that the final tabulations be in any particular form or book. Once the census figures are broken down into counties, towns, wards, etc., and are released to the public by an authorized official, the census is complete for all practical purposes.” Id.
The principle derived from all of these cases is simply that “a new census should be given effect when preliminary figures are made public by those authorized to disseminate such figures.” City of Elyria, 187 N.E.2d at 35; see also State ex rel. Clark, 148 Kan. 632, 84 P.2d 507 (1938). Preliminary census figures filed three months prior to án election, by the enumerator authorized to file them, governed the county’s power to hold that election. Nelson v. Edwards, 55 Tex. 389, 392 (1881).
Finally, the majority hangs its hat, as did Aberdeen, on the caveat that preliminary “population counts are subject to correction for undereount or overcount.” This caveat is appended to every 1990 census report to comply with a stipulation resulting from litigation against the Bureau by some of the nation’s largest cities. City of New York v. U.S. Dept. of Commerce, 739 F.Supp. 761 (E.D.N.Y.1990) (rev’d on other grounds in Wisconsin v. City of New York, — U.S. -, 116 S.Ct. 1091, 134 L.Ed.2d 167 (1996)); see State of New Jersey v. Apportionment Com’n, 125 N.J. 375, 593 A.2d 710, 711 (N.J.1991). The majority claims that this caveat requires that a ten and one-half-year-old census should remain the controlling “latest” census until at least April 1, 1991, or at most until July 15, 1991, the date when the census becomes “official.” This same argument was raised and soundly rejected in State of New Jersey in which the Court held that even where the New Jersey Constitution used the phrase, “official decennial census,” the figures released to the Governor months before either the July 15th or April 1, 1991, deadlines should be considered official and be acted upon by state officers. State of New Jersey, 125 N.J. at 382, 593 A.2d at 714.
The Oklahoma Supreme Court, in Herndon v. Excise Bd. of Garfield County, 147 Okla. 126, 295 P. 223 (Okla.1931), considered a ease where the City of Enid, according to preliminary census figures released on May 3, 1930, was entitled to and did establish a city court. The governing board, claiming that the 1920 census controlled and that the 1930 census was “not complete, nor officially declared,” refused to pay the salary of the new judge. Id., 295 P. 223. The preliminary announcement of the 1930 census figures contained a caveat that “these figures are preliminary and subject to correction.” Id. The initial figures were subsequently corrected after May 3, 1930, but the preliminary announcement was one the Board was required to take notice of and be guided by. Id. at 224.
Preliminary census figures still subject to correction had similar force and effect in Ludwig v. Board of County Comr’s of Sarpy County, 170 Neb. 600, 103 N.W.2d 838 (Neb. 1960); Excise Bd. of Washita County v. Lowden, 189 Okla. 286, 116 P.2d 700 (1941); Elliott v. State ex ret Kirkpatrick, 150 Okla. 275, 1 P.2d 370 (Okla.1931); and Holcomb v. Spikes, 232 S.W. 891 (Tex.Civ.App.1921).
In the case at bar, MIss.Code Ann. § 67-1-14 (Supp.1990) did not require that the census be “official,” only that it be the “latest” census governing population count in Aberdeen for purposes of a city holding a local option liquor referendum. Therefore, the 1990 census should be controlling under § 67-1-14 which became effective June 2, 1990, after the 1990 decennial census date.
*214Even considering the caveat, the Census Bureau, on July 22, 1991, announced that no adjustment to Aberdeen’s preliminary population figures, made public by publication during August, September, and October of 1990, would be made to the final “official” figures. This was so in spite of the Bureau’s admission that two per-cent of the population was not counted.
I respectfully dissent.
PRATHER, P.J., and JAMES L. ROBERTS, Jr., J., join this opinion.