Group Title: Working paper - International Agricultural Trade and Policy Center. University of Florida ; WPTC 05-04
Title: Price transmission and food scares in the U.S. beef sector
Full Citation
Permanent Link:
 Material Information
Title: Price transmission and food scares in the U.S. beef sector
Series Title: Working paper - International Agricultural Trade and Policy Center. University of Florida ; WPTC 05-04
Physical Description: Book
Language: English
Creator: Livanis, Grigorios
Moss, Charles B.
Publisher: International Agricultural Trade and Policy Center. University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: June 2005
 Record Information
Bibliographic ID: UF00089801
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.


This item has the following downloads:

wp050004 ( PDF )

Full Text

WPTC 05-04

I '-ional Agricultural Trade and Policy Center

Grigorios Livanis & Charles B. Moss

WPTC 05-04 June 2005





Institute of Food and Agricultural Sciences



The International Agricultural Trade and Policy Center (IATPC) was established in 1990
in the Institute of Food and Agriculture Sciences (IFAS) at the University of Florida
(UF). The mission of the Center is to conduct a multi-disciplinary research, education and
outreach program with a major focus on issues that influence competitiveness of specialty
crop agriculture in support of consumers, industry, resource owners and policy makers.
The Center facilitates collaborative research, education and outreach programs across
colleges of the university, with other universities and with state, national and
international organizations. The Center's objectives are to:

* Serve as the University-wide focal point for research on international trade,
domestic and foreign legal and policy issues influencing specialty crop agriculture.
* Support initiatives that enable a better understanding of state, U.S. and international
policy issues impacting the competitiveness of specialty crops locally, nationally,
and internationally.
* Serve as a nation-wide resource for research on public policy issues concerning
specialty crops.
* Disseminate research results to, and interact with, policymakers; research, business,
industry, and resource groups; and state, federal, and international agencies to
facilitate the policy debate on specialty crop issues.

Price Transmission and Food Scares in the U.S. Beef Sector

Grigorios Livanis (University ofFlorida) and Charles B. Moss (University ofFlorida)

Presented at the American Agricultural Economics Association Meetings
in Providence, Rhode Island 2005

Copyright 2005 by Grigorios Livanis and Charles B. Moss. All rights
reserved. Readers may make verbatim copies of this document for non-commercial
purposes by any means, provided that this copyright notice appears on all such copies.

Price Transmission and Food Scares in the U.S. Beef Sector

Grigorios Livanis (University ofFlorida) and Charles B. Moss (University ofFlorida)

Abstract: The advent of mad cow disease in Canada in the United States raises numerous

concerns regarding consumer reaction to information in the United States. To examine the role of

consumer reaction to information we examine the response of price spreads in the U.S. beef

market to a Food Safety Index derived from Lexis-Nexis. Specifically, we estimate a vector error

correction model to examine the long and short-run effect of news on price spreads. Our results

indicate that informational shocks are fairly transient in the retail prices, but persist at the

wholesale and farm level.

Keywords: vector error correction, food safety, price spread, beef markets

The incidence of foodborne diseases has dramatically increased in the past fifteen years in the

United States and in other industrialized countries, partially as a result of better surveillance

programs or by the appearance of new pathogens. According to the Centers for Disease Control

and Prevention (CDC) the incidence of food borne disease outbreaks in the United States was in

the range of 500-800 for the period 1990-1997, while in the period 1998-2001 (CDC initiated

improved reporting in 1998) increased to 1026-1380. Parallel to this increase, widely publicized

outbreaks have led to greater consumer awareness about potential food hazards and demand for

safer products. Even more it has led, in some cases, to disruption of international trade for food

and agricultural products.

The impact of food safety information and in particular food scares, as well as product

recall information, on consumer demand for food and agricultural markets has been extensively

analyzed in the literature. These studies can be categorized into three general approaches. The

first set of studies use direct elicitation of consumer preferences methods such as contingent

valuation, conjoint analysis and experimental auction to value food safety information and

recover stated and/or revealed preferences of the consumer for food safety. For instance,

Dickinson and Bailey (2002) found that consumers in the state of Utah are willing to pay the cost

for a greater transparency in red-meat products. Another set of studies focuses on consumer

response to food borne disease outbreaks providing various explanations for either short-run or

long-run consumer behavior. Most notably, Jin and Koo (2003) and Peterson and Chen (2003) in

their analysis for Japanese consumers and Mangen and Burrell (2001) for Dutch consumers have

found that consumers' tastes for meat have systematically moved away from beef to its

substitutes following the discovery of BSE. Finally the third set of studies focuses on the impact

of food safety information on the demand for the risky good. Food safety information is captured

through an index based on media articles or television publicity of food borne disease outbreaks.

For instance, Piggott and March (2004) analyzed the impact of food safety information on US

meat demand and found that the average demand response to food safety concerns was small for

the period 1976-1999. Finally, Huang et al. (2004) after eliciting through surveys consumers'

perceived risks, they estimated the impact of those risks to seafood consumption in North


Another extensive strand of literature has examined price linkages among farm,

wholesale, and retail markets for meat and livestock products in several countries (Palaskas

1995, von Cramon-Taubadel 1998, Goodwin and Holt 1999, Tiffin and Dawson 2000, Abdulai

2002). The general result of these studies is that the transmission between farm, wholesale and

retail prices is asymmetric, whereas there is a unidirectional price transmission from producers to

retailers. Further, a number of these studies could not find a long-run relationship for meat

prices, especially for the United Kingdom.

Although these studies provide many significant results and insights, there has been little

analysis of the impact of food scares on prices at different stages of the beef marketing chain in

the United States. Such an analysis becomes essential if we notice that, in spite of increased

consumer awareness to food safety, real prices in the retail U.S. beef sector have fallen by only

5% (on average) during the period 1990-2004, while in the wholesale and farm sector have

decreased by 38% and 37% respectively. In support of this result, during this period all price

spreads have been observed to grow, but the retail-wholesale price spread has grown five times

more than the wholesale-producer spread (on average). A possible explanation of this outcome

can be attributed to an increased industry concentration and exertion of market power, given the

mergers and acquisitions in the meatpacking industry. Further, the U.S. livestock sector has

experienced several structural changes, which apart from the industry concentration are also due

to changes in marketing practices, increase in the scale of operations, and regulatory policies,

such as the implementation of HACCP from 1997.

Therefore, this study will examine the impact of food scares on U.S. beef prices at

different stages of the marketing chain for the period 1990-2004. Goodwin and Holt (1999)

examined price transmission of shocks in the producer, wholesale and retail marketing channels

in U.S. beef market using a threshold error correction model, and evaluated the dynamic time

paths of price adjustments to shocks at each level in the U.S. beef marketing channel. However,

their approach does not take into account exogenous demand shocks, such as the increased

awareness of consumers in food safety.

Econometric Methods

The methodological approach that this study uses is similar to the study by Lloyd et al. (2001)

who showed that food scares in the United Kingdom resulted in decline of prices at each

marketing stage. Moreover, these price declines were not equal between stages suggesting that

the UK food chain is characterized by some degree of market power. The formal analysis of the

present study is conducted in a cointegration framework, where beef prices at each level of the

marketing chain are endogenous. To capture the increased awareness of consumers to food

safety, we follow Piggott and Marsh (2004) by constructing a food safety index based on

newspaper articles from the popular press in the period 1990-2004. Since the food safety index is

an exogenous shock to beef demand, we assume that changes in this index cause price changes

and not the reverse. Tests include unit root of each time series and a system cointegration test

where we test for unknown structural breaks in each series, using recently developed techniques

in the literature. The latter constitutes the departure from the Lloyd et al. study and a significant

addition to the literature. We follow the studies by Lutkepohl et al. (2004) to test for unknown

structural breaks in the cointegrating relationship; and Saikkonen and Lutkepohl (2002) for the

unit roots. Further, we derive the impulse response functions of the three prices to a unit shock of

the food safety index.

To investigate the order of integration I(.) of the four individual series we use the

Augmented Dickey Fuller (ADF) test. However, if there is a level shift in the level of the data

generating process (DGP), then the ADF test may be distorted. Thus, we use the tests of

Saikkonen and Lutkepohl (2002), Lanne, Lutkepohl and Saikkonen (2002, 2003) to first search

for a structural break in each individual series and then test for a unit root. To illustrate this test

we consider a special case of Saikkonen and Lutkepohl (2002). Thus, let an individual time

series x,, (t = 1,..., T) be generated by the following mechanism

x, = + t + 6d, +z, (1)

where d, is a simple shift dummy d,= 0 if tT,}, (/o,,/,) are

deterministic terms and z, is an unobservable stochastic error generated by an AR(p) process

with possible unit root. To test for a unit root Saikkonen and Lutkepohl (2002) propose to

estimate the deterministic part first by a generalized least squares procedure and then subtract it

from the original series. Then an ADF test can be performed on the adjusted series

z, = x, /, /tt 8d,. Critical values are reported in Lanne, Lutkepohl and Saikkonen (2002).

The cointegration analysis applied in this study follows the approach developed by

Johansen (1988) and Johansen and Juselius (1990). Following a more extensive discussion of the

model presented in Johansen (1995), the estimation procedure is based on the error correction


Ax, = nxt + Fix, + -- + rFp ,_ l + YD, + v, (2)

where x, is the vector of jointly determined I(1) endogenous variables, Ax, is the difference in

x,, D, is a vector of deterministic and/or exogenous variables, F,...F--k, n, and Y are

estimated parameters, and v, is a vector of n.i.d. disturbances with zero mean and non-diagonal

covariance matrix Y. In this study, the endogenous variables are the natural logarithm of beef

prices at various levels of the market channel and the food safety index, whereas the

deterministic variables are impulse dummies and a constant term. The endogenous variables are

cointegrated if the H matrix is singular ( = o'P3). The number of cointegrating vectors is

determined by the significance of the eigenvalues of H or by the trace statistic. Cheung and Lai

(1993) suggest that the trace test shows more robustness to both skewness and excess kyrtosis in

the disturbances than the maximum eigenvalue. Following this result the choice of the rank r

will be based on the trace statistic. Within this formulation the 3 matrix contains the long-run

equilibria, while the ac matrix depicts the speed of adjustment toward the long-run equilibria.

To investigate the impact of changes in the food safety index, in the marketing chain

prices we will perform impulse response analysis. Thus, consider the level representation of (2),

which is a vector autoregressive process VAR(p):

x,= Ix, I + xt-2 +' + p X + YDt + v, (3)

Lutkepohl and Reimers (1992) showed that the impulse response function can be found

by imposing a recursive structure on the moving average representation of the VAR (equation 3)

, =( ((p,)= s- kAk where Q0 =I, Ak =0 for k> p. The impulse response function of

variable i with respect to a unit shock to variable j, s periods ago, everything else constant is

given by a plot of p s. In this analysis we report orthogonal impulse responses but notice that

the results depend on the order in which the variables appear in the VAR.

Data and Results

Our empirical model analyzes three series of monthly beef prices and one series of a food safety

index from January 1990 to December 2004, giving a total of 180 observations. All price data

were obtained from the Economic Research Service of the USDA. The food safety index is

constructed based on newspaper articles from the popular press. Data for this index were

obtained by searching the top fifty English language newspapers using the academic version of

the Lexis-Nexis search tool. Keywords searched were food safety or food contamination or beef

outbreak. Then the search was narrowed only to information about beef by using the additional

term beef

Figure 1 presents the monthly food safety index. Some information on the outbreaks and

regulatory policies follows. In December 1995 there was the first suspicion about potential link

of BSE to CJD, which was confirmed in March of 1996. In July of 1996 USDA announced the

Pathogen reduction and Hazard Analysis and Critical Control Points (HACCP) regulation.

According to this rule all establishments would be required to enforce sanitation standard

operating procedures by January 27 of 1997, as well as E. coli process control testing. The

HACCP regulations and provisions of the rule were applicable to large establishments (500 or

more employees) in January 26, 1998 and gradually until 2000 for the smaller establishments. In

August of 1997, the largest meat recall in U.S. history as Hudson Foods Company linked to the

outbreak of E. co/i-tainted hamburgers and agreed to put off the market and destroy 25 million

pounds of ground beef. In June 1999 U.S. banned beef imports from Europe, while dioxins were

found in milk in Germany. In December, 2000 Europe bans beef over 30 months from the U.K.

while in U.S. there was another meat recall due to E. coli (1 million pounds of ground beef). In

July 2002 ConAgra in U.S. recalled 354,200 pounds of ground beef and 19 million pounds of

beef trim and frozen, fresh ground beef due to contamination by E. coli. Finally, in March 2003

Canada had its first case of BSE, while U.S. had its first case in December of 2003 from a cow

that came from Canada.

Figure 2 shows monthly net farm, wholesale, and retail values for beef, where they have

been corrected for inflation by the consumer price index (CPI). The general trend after correcting

for inflation is downward until the end of 1996, while from January of 1997 there is an upward

trend in the series and especially in the net farm and wholesale prices. Therefore, the prices that

consumers pay for beef have increased less rapidly than inflation until the end of 1996. In other

words, the real cost of beef was declining. However, from January 1997, which coincides with

the enforcement of the sanitation standard operating procedures in all establishments in U.S., it is

observed that all prices were increasing; most probably caused by the implementation cost of the


Transforming all price series and the food safety index in natural logarithm form, we test

each series individually for a possible structural break using the method of Saikkonen and

Lutkepohl (2002), after correcting for autocorrelation. We found no statistical significant

structural break for the retail price series and so we performed an ADF test for unit root1. Table 1

indicates that retail price is integrated of order one I(1), where two lags orders were included

according to the Akaike information criterion (AIC) and Hannan-Quinn (HQ) criterion, as well

two deterministic terms: a constant and a trend. Once the series is difference then the ADF test

with a constant but no trend indicates that the series is stationary. As Table 2 indicates, there was

a structural break in each of the other series. Specifically, the natural logarithm of the food safety

index, has an impulse dummy break at 1996: M3 (dummy is 1 in this month and zero otherwise)

as it was expected from visual observation of Figure 1. To test for the order of integration we

used the Saikkonen and Lutkepohl (2002) test, which clearly shows that the safety index has a

unit root for a lag order of three. The producer and wholesale prices both had a structural break

at 1996: M11, which is before the start of the sanitation standards regulation. The break in both

prices is captured with a shift dummy that is one after the break and zero otherwise. An analysis

of the first differences of the two variables rejects unit roots in these series. Taking into

consideration the results on the levels of those two variables it is evident that the variables are

1 Most of the analysis was performed in the JMulTi program developed by Lutkepohl and Kratzig, and is available
online at

well modeled as I(1). The first differences of the variables do not have a level shift anymore but

just an outlier in November of 1996, which is captured by using an impulse dummy variable in

these tests.

The next step of the analysis is to investigate the number of cointegrating relations

between the series. Assuming a constant and a trend in the cointegrating relationships, and using

the Johansen's Trace Statistic in Table 3 and the published statistics in Case II Table B.10 in

Hamilton, we conclude that there are three cointegrating vectors. To check the robustness of this

result we allowed for a shift dummy in 1996:M11 and an impulse dummy in 1996:M3, along

with a constant and a trend. To test for the cointegrating rank in this case, we used the test of

Saikkonen and Lutkepohl (2000), which is presented in Table 4. We conclude that the data

exhibits three cointegrating relationships. Thus, as in Lloyd et al. (2001) for the U.K. the food

safety index is plays an important role in the long-run relationship of beef prices in the United


Using the previous results, we consider a VECM (equation 2) for the four-dimensional

series with cointegrating rank three and two lagged differences. Moreover, the shift dummy is

included in difference form and so it becomes an impulse dummy. Thus, the deterministic terms

in the VECM are the impulse dummies for 1996 (M3 and M11) and a constant term. A linear

trend term was added initially but it was found to be insignificant. The resulting maximum

likelihood estimator of the cointegrating vector is presented in Table 5.

The orthogonalized impulse response functions of the three beef prices to a one percent

shock in the food safety index are presented in Figure 3. It is evident that the producer and the

wholesale price immediately fall by the same amount from increased consumer awareness about

food safety, as captured by the index. Instead, the retail prices initially increase for a short period

of time but then fall. The recovery of each series to its pre-shock level is faster for the retail

prices with the producer prices taking the most time. In the long-run though, all prices increase

as is evident also from Figure 2.

These results indicate that the food safety index is important in determining the long-run

relationship of the beef prices. Increased food-safety concerns lead to a drop in the prices at all

stages in the marketing chain but the recovery period from the shock differs across stages. It is

evident that retail prices are the least affected from increased food scares in the short-run, while

producer and wholesale prices suffer the most. The importance of food safety regulations such as

the HACCP is also evident in Figure 2, since if the food safety concern was repeated and

consumers were not assured about the quality of the product then a shock in the food safety

index could have caused a decrease in all prices in the long-run.

Summary and Conclusions

This paper analyzed the effect of food-safety concerns on beef prices at the producer, wholesale

and retail level. We find that these concerns affect the long-run equilibrium of the beef prices at

each stage of the supply chain. In particular, we captured the awareness of the consumers

towards food safety by constructing an index based on popular press articles on food

contamination concentrating on beef. We show that this index was cointegrated with beef prices

and the long-run relationship was found using a VECM. Moreover, we show that a shock in the

system caused by one percent change in the food safety index has different effects on each price

series. Specifically, retail prices were the less respondent to this shock, while producer prices had

the longer recovery period. The importance of food safety regulations such as the HACCP was

also demonstrated by the fact that all prices in the long-run were increasing after an exogenous

shock in the demand. This may have also implications about the mandatory country-of-origin

labeling for beef products that is effective in 2006. If country of origin labeling increases

consumers' awareness about the safety of the domestic products, its implementation cost will be

absorbed by the consumer in the long-run. This pattern was observed for the introduction of

HACCP, where we observed real prices at all stages of the marketing chain to increase from

1997 and onwards.

Finally, the mad cow case in the United States could not be sufficiently captured by the

dataset, for instance as a potential structural break in the cointegrating relationship, because of

the short period of data available after its incidence.


Abdulai, A. "Using threshold cointegration to estimate asymmetric price transmission in the

Swiss pork market" Applied Economics 34(2002): 679-687.

Dickinson, D. L. and Dee Von Bailey, "Meat Traceability: Are US Consumers Willing to Pay for

It?" Journal ofAgricultural and Resource Economics 27(2)(2002): 348-64.

Goodwin, B. and M. Holt "Price Transmission and Asymmetric Adjustment in the US Beef

Sector," American Journal ofAgricultural Economics 81(1999): 630-7.

Huang, J.C., T.C. Haab, and J.C. Whitehead. "Risk Valuation in the Presence of Risky

Substitutes: An Application to Demand for Seafood." Journal of Agricultural and

Applied Economics 36(2004): 213-28.

Jin, H.J. and W.W. Koo. "The Effect of the BSE Outbreak in Japan on Consumers' Preferences."

European Review ofAgricultural Economics 30(2)(2003): 173-92.

Lanne, M., Lutkepohl, H. and P. Saikkonen. "Comparison of Unit Root Tests for Time Series

with Level Shifts." Journal of Time Series Analysis 23(2002):667-685.

Lanne, M., Lutkepohl, H. and P. Saikkonen. "Test Procedures for Unit Roots in Time Series with

Level Shifts at Unknown Time." Oxford Bulletin of Economics and Statistics 65(2003):


Lloyd, T., McCorriston, S., Morgan, C. W. and A. J. Rayner. "The Impact of Food Scares on

Beef and Inter-Related Meat Markets." American Agricultural Economic Association

Annual Meeting, Selected paper, Chicago, IL, August 2001.

Lutkepohl, H., P. Saikkonen and C. Trenkler. "Testing for the Cointegrating Rank of A VAR

Process with Level Shift at Unknown Time." Econometrica 72(2)(2004): 647-62.

Mangen, M.J. and A.M. Burrell. "Decomposing Preference Shifts for Meat and Fish in the

Netherlands." Journal of Agricultural Economics 52-2 (2001): 16-28.

Peterson, H. H. and Y. Chen. "The Impact of BSE on Japanese Retail Beef Market." Selected

Paper at the Southern Agricultural Economics Association Annual Meeting, 2003.

Palaskas, S. "Statistical Analysis of Price Transmission in the European Union." Journal of

Agricultural Economics 41(1995): 61-69.

Piggott, N. E. and T. L. Marsh. "Does Food Safety Information Impact U.S. Meat Demand?"

American Journal ofAgricultural Economics 86(1)(2004): 154-74.

Saikkonen, P. and H. Lutkepohl. "Testing for a Unit Root in a Time Series with a Level Shift at

Unknown Time," Econometric Theory 18(2002): 313-48.

Tiffin R, and P.J. Dawson "Structural Breaks, Cointegration and the Farm Retail Price Spread for

Lamb." Applied Economics 32(2000): 1281-1286.

Von Cramon-Taubadel, S. "Estimating Asymmetric Price Transmission with the Error-

Correction Representation: An Application to the German Pork Market." European

Review ofAgricultural Economics (1998): 1-18.

Saikkonen, P. and H. Lutkepohl. "Testing for the Cointegrating Rank of a VAR Process with

Structural Shifts, Journal of Business & Economic Statistics 18(2000): 451-464.

Centers for Disease Control and Prevention (CDC). December 1999. Internet address:

htto://www.cdc. ov/foodborneoutbreaks/outbreak satistics/cdc renorted.htm.

Figure 1. Food Safety Index









00 (NI ] n rl- O xO N N 0 Q o 00 0 C en en

c 3^ 3^"3^" il'^ f3 ~'''I'3 ^'3'' f 3^' 3^'3^' 3^' '3
0 > 0 > 0 > 0> 0 > 0 0 0 0 0 000000000
I I, I, I, I, I, I, I, I, I, I, I, I, I, I I I I

Figure 2. Real Monthly Beef Prices











o CC C "D rm- 00 00ooo cC C oo C)NNC m M -t

-- Net Farm Wholesale Retail

Figure 3. The Dynamic Effect of Shocks to the Food Safety Index



0 0.002

) -0.002

| -0.004


0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Months After Shock

-- Net Farm Price -- Wholesale Price -- Retail Price

Table 1. ADF Tests for Retail Price

Variable Levels (lag) 5% critical value Differences (lag) 5% critical value

retail price -0.6658(2) -3.41 -10.52 (1) -2.86

Notes: Lag length for the ADF regression selected according to the Akaike Information Criterion (AIC) and
Hannan-Quinn Criterion and are reported in parentheses adjacent to the test statistics. The levels regression
includes a constant and trend and the differences includes only a constant.
Critical values from Davidson and MacKinnon (1993).

Table 2. Unit Root Tests in the Presence of Structural Shifts for Net farm, Wholesale Prices
and Food Safety Index
5% critical 5% critical
Variable Break Date Shift function Levels (lag) value Differences (lag) value
value value
farmp 1996: M11 Shift dummy -1.95 (3) -3.03 -9.40 (2) -2.88
wholesp 1996: M11 Shift dummy -1.78 (3) -3.03 -10.61 (2) -2.88
FSI 1996: M3 Impulse dummy -2.64 (3) -3.03 -11.96(2) -2.88
Notes: Lag length for the unit root regression selected according to the Akaike Information Criterion (AIC)
and Hannan-Quinn Criterion and are reported in parentheses adjacent to the test statistics. The levels
regression includes a constant and trend and the differences includes only a constant. Notice that
in differences the shift dummies become impulse dummies.
Critical values from Lanne, Saikkonen and Lutkepohl (2002).

Table 3. Cointegration Test Statistics
Ho LR Trace Statistic 5% Critical Values
r = 0 115.05 63.66
r = 1 67.29 42.77
r =2 30.96 25.73
r = 3 4.06 12.45
Note: This test is based on Johansen's procedure (1991).

Table 4. Cointegration Test Statistics with Shift and
Impulse Dummies
Ho LR Trace Statistic 5% Critical Values
r = 0 115.05 63.66
r = 1 67.29 42.77
r =2 30.96 25.73
r = 3 4.06 12.45
Note: This test is based on Saikkonen and Lutkepohl (2000).

Table 5. Cointegrating Vector
Variable Vector 1 Vector 2 Vector 3
Farm Price 1.000 0.000 0.000
(0.000)a (0.000) (0.000)
Wholesale Price 0.000 1.000 0.000
(0.000) (0.000) (0.000)
Retail Price 0.000 0.000 1.000
(0.000) (0.000) (0.000)
Food Safety Index 0.354 0.332 0.313
(0.062) (0.062) (0.072)
Constant -6.597 -6.677 -7.145
(0.253) (0.250) (0.291)
aNumbers in parenthesis denote standard deviations

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs