A collaborative investigation of phages for typing bovine staphylococci

Bull. Org. mond. Sante 1972, 46, 81-98Bull. Wld Hith Org f A collaborative investigation of phages for typing bovine staphylococci IAN DAVIDSON' on behalf of the collaborators Seventeen collaborators tested a total of6 999 cultures. The results were used to assess the usefulness of each phage for typing bovine staphylococci and to classify the phages into lytic groups and the cultures into phage groups. Some marked differences in the distri- bution of phage patterns from country to country were found. Evidence concerning the stability of the phage patterns is presented. On the basis of the results, a set ofphages that would be generally useful for typing bovine staphylococci was selected. This set was established as the international basic set ofphages for typing bovine staphylococci by the International Subcommittee en the Phage Typing of Staphylococci. The international phage set that was selected for typing human coagulase-positive staphylococci 8 has proved inadequate for typing staphylococci from other animals. Workers interested in typing bovine staphylococci have, consequently, selected different phages, including phages isolated from or adapted to bovine cultures (Bertoni & Rosaschino, 1963; Bonin & Blobel, 1967; Coles & Eisenstark, 1959; Davidson, 1961; Frost, 1967; Fujikura & Shibata, 1965; Gedek, 1966; Meyer, 1967; Mondini & Dova- dola, 1959; Nakagawa, 1960; Seto & Wilson, 1958; Smith, 1948; Verge et al., 1960). In order to facilitate the comparison of results and to make the most useful phages generally available, the author submitted proposals to the International Subcommittee on Phage Typing of Staphylococci of the International Association of Microbiological Societies for the development of an international phage set for bovine staphylococci (Int. Bull. bact. Nomencl., 1963, 13, 119; Int. J. system. Bact., 1967, 17, 113). These proposals were accepted and all workers known to be active in the field were invited 1 Ministry of Agriculture, Fisheries and Food, Central Veterinary Laboratory, New Haw, Weybridge, Surrey, England. 'Described in the reports of the International Subcom- mittee on Phage Typing of Staphylococci (Int. Bull. bact. Nomencl., 1959, 9, 115; Int. Bull. bact. Nomencl., 1963, 13, 119; Int. J. system. Bact., 1967,17, 113). to take part in an investigation of a preliminary selection of phages that showed promise of being generally useful. Collaborators were asked to use, as a minimum, the following phages: international phages: 29, 52A, 3A, 6, 53, 75, 77, 84, 42D. bovine and other phages: 78, 102, 107, 1363/14, SI, S6, 883, ACI. If possible, all the phages of the international basic set for human staphylococci were to be used. In addition, collaborators were asked to test the stability of the phage patterns of representative cultures. Each collaborator obtained seed material for the propa- gation of the bovine phages from Weybridge and samples of the international phages from his natio- nal representative on the International Subcom- mittee. MATERIALS AND METHODS Data submitted The 17 collaborators listed in the Annex submitted details of the phage reactions they obtained with a total of 6 999 cultures of coagulase-positive staphylo- cocci. Fourteen of the collaborators used most or all of the international phages and 3 used the more 2784 -81 I. DAVIDSON Table 1. Data used in computer analysis 1. Australia 2. Bulgaria 3. Denmark 4. Finland 5. Federal Republic of Germany 113 812 1 344 1 034 Sources of cultures Phages used a more than 288 herds in 7 districts: normal and mastitis milk set A except 85 and 187. Additiona phages: 3B, 7,42DA, 101, 367, 425 600,10,186,373 milk of individual cows 494 cultures from churn milk samples from 223 herds: 318 cultures from milk of individual cows from 9 herds 650 herds: normal and mastitis milk more than 40 herds from the north and south of the country: milk of individual cows set A set A except 85. Additional phages: 38, 7, 42F set A except 52, 79, 85, and 187 set A except 80, 83A, and 85. Additional phages: 7, 825 6. Federal Republic 200 200 herds: milk of individual cows set B except 78 of Germany 7. German Democratic 249 91 cultures from Czechoslovakia from mastitis milk, remain- set A. Additional phages: 42F, Republic ing cultures from 3 districts of the GDR: 124 from cows' 108, 111, 812, 825 milk and 34 from dairy products 8. Ireland 128 91 cultures from bulk milk of different herds: 37 cultures set B from milk of individual cows of one herd 9. Israel 97 31 herds: normal and mastitis milk set B 10. Japan 300 northern and central Japan: milk from individual cows set A except 85. Additional phages: 3B and 7 11. New Zealand 190 several districts: 10 cultures from cheese, 2 from calves, set A except 85. Additional phages: remainder from normal and mastitis milk 38, 7, 101, 105, 108, 110, 111, A13 12. Sweden 226 61 herds: mastitis milk set A 13. United Kingdom 454 more than 300 herds from all parts of the UK: normal and set A. Additional phages: 3B, 7 mastitis milk 129/16, P42D/E193, 88A, 111, H98, 365, T90, 257 14. USSR A 120 3 herds: 10 cultures from milkers' hands, 22 from milking set A, except 80, 3C, 47, 54, 84, apparatus, the remainder from normal and mastitis milk and 85. Additional phages: 3B and 7 15. USSR B 180 98 cultures from mastitis milk, 82 from dairy products set A, except 85. Additional phages: 3B and 7 16. USA A 442 1 herd sampled over 3 years: milk from individual cows set A except 84 and 85. Additio- nal ph3ge: 3B 17. USA B 237 more than 50 herds in 12 States, 9 cultures from Canada: set A except 84, 85 and 78. normal and mastitis milk Additional phage: S2 a See Table 2. 82 ~~I I PHAGES FOR TYPING BOVINE STAPHYLOCOCCI Table 2. Phages used in computer analysis Set A Current international set and bovine phages from provisional bovine set: 29, 52, 52A, 79, 80, 3A, 3C, 55, 71, 6, 42E, 47, 53, 54, 75, 77, 83A, 84, 85, 42D, 81, 187, 78, 102, 107, 1363/14, Si, S6, 883, AC1 Set B Provisional bovine set: 29, 52A, 3A, 6, 53, 75, 77, 84, 42D, 78, 102, 107, 1363/14, S1, S6, 883, AC1 Additional phages 3B, 7, 42F - phages of international series not in basic international set. 129/16, P42D/El 93, 88A - Smith, H. W. (1948) S2- Seto, J. T. & Wilson, J. B. (1958) Al 3 - Coles, E. H. & Eisenstark, A. (1959) H98, 365, T90, 257- Nakagawa, M. (1960) 101, 105, 108, 110, 111 - Davidson, I. (1961) 812, 825- Meyer, W. (1966) 367, 425, 600, 10, 186, 373 - Frost, A. J. (1967) 42DA - a variant of phage 42D obtained in Australia limited selection listed above. Details are given in Tables 1 and 2. Table I also gives the sources of the cultures used. Most of the cultures were obtained from cow's milk but some came from dairy products, milking appa- ratus, calves, and human contacts. Some collabo- rators were able to obtain cultures from a large num- ber of different herds in different districts while others tested a more limited selection. Nine collaborators used the phages at RTD (rou- tine test dilution) and at 1 000 RTD (or in one case, undiluted). Eight collaborators used the phages at RTD only. Nine collaborators conducted studies of the stability of phage patterns. Twelve colla- borators submitted the results of their lytic spectra determinations and/or of their determinations of the phage patterns of the propagating strains. Two labo- ratories submitted, in addition, the results of testing cultures with other selections of phages. Analysis ofphage reactions To assist in analysing the phage reactions of the 6 999 cultures, a computer programme was used to analyse the results of each collaborator individually and to analyse the combined results. Since not all collaborators used all the phages, the combined results were analysed in three stages. First, the results of those collaborators using all or almost all the phages of set A (see Table 2) were added together. The collaborators concerned (group 1) are no. 1, 2, 3, 7, 10, 11, 12, 13, 15, and 16. They tested 3 839 cultures. Since six ofthem did not use phage 85, this phage was disregarded in making the analysis. One collaborator in group 1 did not use phage 84 and one did not use phage 187. Next, the results of those collaborators using a rather smaller number of the phages in set A (colla- borators no. 4, 5, and 17) were added to those of group 1 to form group 2 and a total of 6 455 cul- tures. Phage 85 was again disregarded. Phages 52, 79, 80, 83A, and 78 were not used by one collaborator each in this group and phages 84 and 187 were not used by two collaborators. Finally, the results of all collaborators were taken together to form group 3. In performing the analysis, due allowance was made for the fact that not all collaborators used all the phages. The analysis of individual results was carried out at two levels; (a) including only strong reactions (i.e., more than 50 plaques) at RTD and (b) in- cluding all reactions at RTD. The combined results were analysed by the computer at level (a) only, since the information needed from the level (b) analysis could be obtained fairly easily by adding together the individual analyses. Collaborator no. 5 did not record weak reactions with the international phages. In this case, confluent lysis at 1 000 RTD was used as level (b) but it should be noted that cultures that gave strong reactions at RTD were not tested at 1 000 RTD. Collaborator 83 I. DAVIDSON no. 17 recorded only two levels of reaction at RTD, i.e., confluent lysis and 10 plaques to semi-confluent lysis, inclusive. The first level was used as level (a) and the two together as level (b). The successive stages of the analysis are described below. Stages 1-5 inclusive, 7, and 8 were carried out by the computer and the remainder by hand. (1) The computer counted the number of cultures reacting with each phage. (2) For each of the counts at (1), it counted the number of those cultures also reacting with each other phage. (3) The computer printed the counts at (1) and (2) in the form of two triangular matrices of numbers, one matrix for level (a) and one for level (b). Example: Phage no. 29 52 52A 29 173 142 105 52 156 105 52A 112 Thus 173 cultures reacted with phage 29. Of the 173, 142 also reacted with phage 52, etc. (4) The computer calculated the counts at (1) as percentages of the total number of cultures (454 in the example), and the counts at (2) as percentages of the counts at (1). Example: Phage no. 29 52 52A 29 38a 82b 61 b 52 91 C 34 a 67 c 52A 94 d 94 d 25 a a= %of 454 b = % of 173 c = %of 156 see example at (3) d= % ofl12 J (5) The computer examined the results at (4) and printed lists of pairs of phages that showed more than 75%, more than 65%, and more than 50% similarity both ways (i.e., more than 75% of cultures reacting with phage A also reacted with phage B and more than 75% of cultures reacting with phage B also reacted with phage A, etc.). (6) These results were examined by hand and a set of phages was chosen according to the criteria listed below. (7) The computer then calculated the percentage of cultures typed (i.e., giving strong reactions) at RTD by the phages chosen at stage (6). (8) Finallythe computer identifiedcultures not typed by the chosen phages and reexamined their strong re- actions at RTD with the excluded phages in the same way as stages (1)-(3) above. Any useful phages that had been wrongly eliminated could then be identified. (9) The data from stage (4) of the group analyses were used to determine the lytic groups of the phages in the manner of Williams & Rippon (1952). The cultures were then classified, collaborator by colla- borator, according to the lytic groups of the phages that lysed them. (10) A further search for useful phages was made using the data from stage (9). The aim at stages (6), (8), and (10) was to select a set of phages that would type a high proportion of the cultures at RTD and that would make as many reproducible distinctions between cultures as pos- sible. To this end, the following criteria were applied. (1) Phages that raised the percentage of typable cultures significantly were chosen. In the case of a collection of 100 cultures, any phage that typed 4% or more of the cultures not typed by the other chosen phages was selected provided that it satisfied the other criteria. For 200 cultures the figure of 2% was used, for 1 000 cultures, 1 %, and for 5 000 or more cultures, 0.5 %. Preference was given to those phages that raised the percentage of typable cultures by a large amount relative to the total number of cultures they lysed. (2) In addition, phages were chosen that subdivided the cultures lysed by those of the other chosen phages that lysed large numbers of cultures. Such " subdivid- ing" phages were required to lyse between 20% and 50% of cultures sensitive to one of the other chosen phages. Generally not more than one " subdividing" phage was chosen for each other phage unless the latter typed a very large number of cultures. (3) Not more than one of any group of similar phages was chosen. No two phages that showed more than 50% similarity both ways (see stage (5) of com- puter programme) at level (a) or more than 65% simi- larity at level (b) were chosen unless it was necessary to include such phages in order to raise the percen- tage of typable cultures to an acceptable level. 84 PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 85 (4) As far as possible, phages that gave large num- RESULTS bers of weak reactions were excluded. Generally, a phage was excluded if the number of reactions it gave Phage reactions at RTD at level (b) was more than twice the number at level (a). Table 3 shows, for each collaborator, the percen- (5) In choosing between similar phages, preference tage of cultures lysed strongly by each phage at was given to those from the international set so as to RTD. There was considerable variation from one avoid the needless introduction of new phages. collaborator to another. Some of this is no doubt Table 3. Percentage (to the nearest whole number) of each collaborator's cultures giving a strong reaction with each phage at RTD * Collaborator Phage i1 2 [ 3 4 5 6 1 7 8 - 9 10 11 12 - 13 14 15 16 17 1~ 29 5 2 2 9 17 10 0 61 1 6 21 1 38 0 1 2 13 52 7 4 2 13 0 1 23 4 34 1 2 1 11 52A 0 3 2 4 11 2 1 43 4 2 10 3 25 1 1 0 8 79 0 3 0 8 0 1 2 2 2 0 1 0 17 80 1 3 3 6 0 6 8 2 33 1 0 11 3A 1 1 43 0 0 1 1 2 0 0 2 15 2 0 1 0 4 3C 0 1 1 1 1 0 1 3 0 0 0 0 2 55 1 0 0 0 1 2 0 0 0 0 1 0 0 2 71 1 1 0 0 2 0 0 0 0 0 1 0 0 2 6 7 1 1 2 11 4 6 37 6 1 25 8 31 3 2 1 2 42E 30 2 0 6 14 34 1 34 0 37 52 9 0 3 47 5 1 1 1 10 6 3 21 7 30 2 7 1 1 53 3 3 1 1 4 4 4 21 6 4 14 7 7 0 1 1 4 54 13 2 1 2 11 15 1 20 8 30 2 1 3 75 11 2 1 1 9 3 4 43 2 2 22 4 22 6 1 0 3 77 3 1 0 6 0 2 7 1 2 3 7 4 3 1 1 1 1 5 83A 7 0 1 1 6 3 2 0 28 3 1 1 5 84 2 3 1 3 4 1 2 9 5 1 10 4 13 0 85 3 1 3 5 42D 23 13 4 20 5 40 56 12 2 20 22 0 20 68 34 1 7 81 2 0 2 1 12 3 5 19 3 30 0 7 1 5 187 0 0 0 0 0 0 0 0 0 0 0 3 78 1 0 2 3 0 0 0 0 0 15 1 9 0 0 0 102 9 25 5 26 30 67 65 27 26 25 29 0 18 32 61 13 24 107 22 19 3 33 25 73 70 9 9 24 18 5 14 24 50 15 22 1363/14 28 18 5 60 30 59 63 32 32 25 41 1 1 49 67 62 1 1 16 Si 1 4 0 3 3 2 3 9 2 6 3 0 6 2 2 12 9 S6 4 4 15 9 2 2 0 5 1 0 0 1 9 23 2 52 53 883 2 2 22 0 0 1 60 3 0 11 0 23 2 5 0 0 2 AC1 0 9 1 3 0 1 1 1 1 3 1 0 9 4 1 2 5 * A blank space indicates that the phage was not used by that collaborator. ~ ~ ~ ~ ~ ~ ~ ~ ~ .DAVIDSON random variation in the samples of cultures tested, especially when the number of cultures was small. However, there is a tendency for sets of results from the same country and from neighbouring countries to resemble each other more closely. For example, the results from Denmark and Sweden have a high proportion of reactions with phages 3A and 883; both laboratories in the USA found a high incidence of reactions with the phage S6 and the highest inci- dence of reactions with lytic group I phages was found by the collaborators in Ireland and in England. Percentage of cultures typed The percentage of cultures typed with set A or set B at RTD ranged from 40 to 97 (see Table 4). The use of these phages at 000 RTD resulted in a con- siderable rise in the percentage typed when this was low at RTD. The additional phages gave a useful increase in the percentage typed in only two cases. The percentage typed by the international phages alone was always lower and often considerably lower. Lytic groups ofphages The results of the three groups of collaborators (see above) were used to produce correlation charts showing the frequency with which pairs of phages occurred in the phage patterns. In interpreting the charts it has to be borne in mind that certain phages were not used by all the collaborators. In the charts for groups 2 and 3, this resulted in these phages show- ing lower correlations with the phages used by all collaborators than they did in the chart for group 1. Since the three charts were otherwise almost identi- cal, only one is reproduced, that for group 3, i.e., the overall results (see Fig. 1). As has been noted before with bovine staphylo- cocci, the difference between lytic group I and lytic group III was not as great as it is with human staphy- lococci. Phage 42E, in particular, appeared to be intermediate between the two groups. On the other hand, phage 81, which is placed amongst the miscel- laneous phages of the international set, behaved as a group-I1l phage. The group-IV phages 42D, 102, 107, and 1363/14 formed a well-defined group. Phage 883 was highly correlated with phage 3A. Its somewhat lower correlations with the group-IV phages were produced entirely by the results of one collaborator. In the reproducibility tests performed by this collaborator, these correlations with the Table 4. Percentage of cultures typed with various sets of phages Cha- Collaborator racte- ristica 1 2 3 4 5 6 17 8 9 10 11 112 13 14 115 16 17 A 74 40 75 91 68 92 83 97 52 58 81 52 89 73 78 78 85 B 85 98b 85 98 98 88 81 100 85 c 85 54 84 91 NRc 93 86 97 67 82 82 63 91 73 82 81 89 D 7 0 0 2 1 6 2 0 0 0 0 E 56 17 54 39 27 55 70 74 18 36 60 31 65 69 46 11 53 F 63 71 94 78 71 98 73 a A. Percentage of cultures typed (i.e., giving one or more strong reactions) at RTD by phages of set A or set B. B. Percentage of cultures typed at 1 000 RTD by phages of set A or set B. C. Percentage of cultures giving any degree of lysis at RTD with one or more phages of set A or set B. D. Of the cultures not typed at A, percentage typed at RTD by additional phages. E. Percentage of cultures typed at RTD by current international phage set. F. Percentage of cultures typed at 1 000 RTD by current international phage set. See Table 1 for the phages used by each collaborator. b Undiluted phages. c NR=not recorded. 86 PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 87 Total no., of cultures n Phage sensitive a, ¢ o <s:oJ V 4 r en -r cS'-4a n C4 4 O O° C - O , Ls Lr r- 00o n Cl n r- %O -I _< n u) r- r- a0 oo co ' 00 m sH H Hv )0 < 29 725 E E__ ***___ *__ 52 461 I I... _....._i_..i..._ 52A 427 |i * IE Em E .__*_ _ mur- 79 153 _ _ _| - _ _ __ |_ _ 80 323 *** *Eu- _ * * - 3A 425 3C 49 *rn_nr_mfl ***_._ _ _* 55 32 _ _ _ __ __ _ _m._ 71 48 _ _i U E * 6 522 MI_... 42E 897 E * _ 47 462 *f.U* I * __ -rn_ _._ 53 245 *___ .IiE ***E_U** rnf_*n. 54 509 _ __ _ 1 - _- _ _ _ _ _ 75 465 -**_ -____E-. 100 % 77 210 -.--_ E .E._ _ _ 83A 265 *i..Ehm.EIS. =-M m MI__ *__ 84 225 *....-EhilE...*U* _ _ 85 36 SOMM_ __111110011 m| 42D 1151 _ _ - 81 396 ihErnEIIIUIiUU.. ____E.__. 187 7 EEEE - ____ - U *IE_n _ 78 140 *U _ _ Ernr___ 102 1630 - - - - - 107 1701 - - - - - 1363/14 2435 ._* Si 234 * *__ S6 741 - - - - - - - - - - - - - _ 883 464.- - - - ACl 153 * - - kr__ ._r Fig. 1. Analysis of correlations between phages to determine lytic groups: results of all collaborators taken together considering only strong reactions at RTD. The height of the columns indicates the percentage of the total also sensi- tive to each other phage. 88 I. DAVIDSON group-IV phages were no longer evident. Phage 883 Phage groups of cultures can thus be allocated, at least provisionally, to lyticgroup~~ ~IL Thrte ru-Lpae ye o e For this purpose the phages were regarded asgroup I. he o her g o p II hages l sed too few bengrudasflo : cultures for their positions to be assessed. beig grouped as follows: Phages 78, Si, and ACI showed low levels of group I: 29, 52, 52A, 79, 80 correlation with many phages and thus have to be group H: 3A, 3C, 55, 71, 883 considered miscellaneous: 29% of the cultures lysed group M: 3, 325, 7, 83 by phage 78 and 76% of those lysed by phage SI group m: 6, 42E, 47, 53, 54, 75, 77, 83A, 84, 85, 81 were not sensitive to any other phage. group IV: 42D, 102, 107, 1363/14 Phage S6 was not correlated with any other phage. miscellaneous: 187, 78, SI, S6, AC1 Table 5. Number of cultures in each phage group ' Collaborator Phage group 1 2 3 4 5 6 7 8 9 10 11 [ 12 [ 13 14 15 16 171 1~~~~~~~~~~~~~~~~~~~~~~~~ 4 16 375 1 20 I 53 0 8 0 0 1 0 0 0 0 0 2 0 18 0 102 5 206 812 22 75 6 1 50 24 856 3 10 14 4 10 0 57 7 0 2 4 0 5 6 47 14 127 29 14 11 6 36 0 428 2 59 3 0 1 21 I 0 69 4 0 3 11 3 3 329 12 5 2 0 10 0 149 0 0 0 0 0 0 0 0 0 0 2 0 2 16 0 0 5 0 6 I 22 0 0 0 0 0 90 18 6 0 0 0 56 0 0 0 3 42 12 O 0 0 7 0 26 0 29 17 3 O 0 9 8 0 0 0 2 6 4 1 344 11 0341 200 249 128 2 0 0 0 9 1 34 0 1 0 0 0 0 0 0 0 0 0 2 0 0 47 97 10 4 36 0 18 3 78 0 3 3 0 0 O O 0 0 3 0 0 10 0 3 126 4 1 3 0 15 6 43 6 10 1 0 0 0 22 4 0 21 9 0 5 0 0 4 36 5 0 53 0 26 0 24 0 4 0 0 0 0 0 0 0 0 0 0 2 2 108 9 10 2 0 25 8 135 6 33 28 1 0 0 13 3 0 59 45 0 7 0 5 13 52 0 0 0 0 0 0 11 8 0 0 0 5 42 18 0 0 0 0 0 0 33 300 190 226 454 1 120 0 3 O 0 30 10 0 0 23 O 0 2 0 0 0 0 0 39 180 2 6 0 2 2 0 36 7 0 0 0 0 0 20 9 0 0 4 0 0 29 222 5 98 8 29 3 1 7 14 23 9 2 4 5 10 0 2 9 2 6 1 0 2 48 15 36 442 237 * The table is based on strong reactions, i.e., more than 50 plaques, at RTD. See text for groups of phages. M = miscellaneous phages; NT = not typable. Il/Mli III IV IV/M 1/111 I/Ill/M l/lV I/IVl Il/lV/M III/IV III/IV/M l/ll/lV/ I/lI/ly l/lll/lV l/III/IV/M lIlll/lV 78 Si S6 others NT total 18 13 21 1 120 8 254 24 0 0 0 133 5 0 18 2 0 2 0 7 18 227 I I0 0 1 0 27 4 0 0 0 0 0 0 0 0 1 0 0 0 8 68 873 113 PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 89 Table 5 shows the phage patterns of the cultures reason for these differences more fully, each colla- classified according to these lytic groups and Table 6 borator's results that gave sufficient information sh'ows the number of cultures sensitive to one phage were examined to determ-ine whether differences only. in the sources of the cultures might be respon- Differences between collaborators similar to those sible. The results of this examination are summa- already noted in Table 3 are evident. To explore the rized below. Table 6. Number of cultures giving a strong reaction at RTD with only one phage of set A or set B Collaborator Phage 1 2 13 4 5 ] 6 7 18 9 [ 10 I11i 12 113 114 1 5 116 117 29 1 0 0 6 5 1 1 0 6 0 3 0 0 2 0 0 2 0 52 10 0 0 2 0 0 0 0 0 0 0 0 1 52A 1 0 0 1 2 0 0 2 0 0 0 0 0 0 0 0 0 79 0 0 0 0 0 0 0 0 0 0 0 0 2 80 0 0 1 0 0 5 0 1 1 0 0 0 3A 1 0 198 0 0 0 0 0 0 0 0 1 0 0 1 0 0 3C 0 0 0 0 0 0 3 0 0 0 0 0 0 55 0 0 0 0 0 0 1 0 0 0 0 0 0 0 71 2 0 2 0 2 0 0 0 0 0 0 0 0 0 6 0 0 0 3 2 1 0 0 2 0 0 0 1 0 0 0 0 42E 43 0 0 5 0 0 1 1 0 0 0 2 0 0 47 0 0 0 0 1 0 0 0 0 1 0 0 2 53 1 0 2 0 1 2 0 0 1 3 1 0 1 0 0 0 0 54 7 0 0 1 0 1 0 0 6 0 0 0 0 75 10 0 0 1 4 1 0 1 0 1 0 0 0 0 0 0 0 77 0 0 0 18 0 0 0 0 0 0 0 0 0 0 0 1 0 83A 3 0 0 1 0 1 0 0 0 0 0 1 0 84 1 0 3 2 3 1 0 0 2 1 1 0 1 0 85 0 0 0 0 42D 59 2 0 19 7 0 2 0 0 1 1 0 11 1 2 0 1 81 0 0 6 0 2 1 4 0 1 3 0 0 0 1 187 0 0 0 0 0 0 1 0 0 0 0 0 78 2 0 18 5 3 0 0 0 0 5 1 7 0 0 0 102 6 0 0 27 127 2 2 1 1 3 4 3 0 10 0 5 1 3 107 26 0 5 10 46 2 0 0 0 2 3 0 6 0 1 5 0 1363/14 46 0 3 256 1 1 2 0 0 10 0 8 13 37 0 2 8 0 Si 0 0 0 6 1 1 2 0 0 2 10 0 0 0 0 0 29 2 S6 7 0 102 47 3 0 0 2 0 0 0 2 5 0 1 222 48 883 16 0 19 4 0 1 2 0 0 32 0 18 2 0 0 0 2 ACI 3 5 1 2 0 0 0 0 0 3 0 0 0 0 0 2 1 *A blank space indicates that the phage was not used by that collaborator. I. DAVIDSON Collaborator 1. Phage patterns were predomi- nantly III, IV, and IH/IV but there was a high pro- portion of883 and untypable cultures as well as group- III and group-IV cultures from Western Australia. Collaborator 3. Group-II and S6 cultures pre- dominated in both churn samples and milk from indi- vidual cows. There was a greater variety of phage patterns amongst the former but this was to be ex- pected since they represented 223 herds while the individual cow samples represented only 9. Collaborator 5. Cultures from the southern part of the Federal Republic of Germany were mostly group IV. Amongst those from the northern part of the Fed- eral Republic of Germany, however, there were more 1/111/IV and III/IV cultures than group IV cultures. Collaborator 7. No significant differences were found between cultures from the German Demo- cratic Republic and those from Czechoslovakia. A higher proportion of untypable cultures was obtained from dairy products than from fresh milk, possibly because staphylococci from human sources were pre- sent in the former. Collaborator 11. The two cultures from calves and the ten from cheese had phage patterns similar to those of cultures from milk. Collaborator 13. There was little difference be- tween phage patterns of cultures from different areas or between those of cultures from clinical mastitis and herd samples. Collaborator 14. There were no significant diffe- rences between cultures from different sources. Collaborator 15. Group IV and III/IV cultures predominated in cultures from both mastitis milk and dairy products. Collaborator 17. There were no significant diffe- rences between areas except that there was a higher proportion of S6 cultures from three of the four herds sampled in New Hampshire. Thus there is evidence that, in some cases, phage patterns varied from district to district in the same country as well as from country to country. There is no evidence that the way the samples for culture were selected influenced the results. Stability ofphage patterns These tests took three main forms: (1) testing at different times subcultures of the same original cul- ture, (2) testing simultaneously single-colony sub- cultures derived from the same culture, and (3) testing repeated isolations from the same site. Although the tests revealed some instability of phage patterns, most cultures varied only slightly or not at all. No phage was markedly more variable than the others in this respect. The results summarized in the tabulation below refer to tests at RTD unless otherwise stated. The phages used were those listed in Table 1 unless otherwise stated. The results with the additional phages are not recorded below. The figures at A are the numbers of cultures that gave a strong (i.e., + +) reaction on at least one occasion. The figures at B are the numbers of cultures that gave a + + reaction on at least one occasion and no reaction at all on at least one occasion. Collaborator 3 Four cultures were tested 10 times and one culture 9 times over an 18 day period. With one excep- tion, variations were noted only in weak reactions. 29 52 52A 80 3A 3C 55 71 6 42E 47 54 75 83A 84 81 78 102 1363/14 Si S6 883 A 3 3 3 5 11 1 1 1 2 12 1 1 1 2 1 2 2 1 3 2 B 0 0 0 0 0 0 0 0 0 0 00 1 0 0 0 0 0 0 0 0 0 Six single-colony subcultures from each of 6 cultures and 4 from each of 2 cultures were tested. Three of these single-colony subcultures showed some variation and a further series of 3 or 4 single-colony sub- cultures from each was tested. Six of these 8 cultures varied only in their weak reactions. nn PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 29 52 52A 79 80 3A 3C 55 71 6 42E 47 54 RT A 1 211 1B 0 0 0 0 0 RTD+1 000 RTD A 2 1 3 1 2 4 2 2 2 4 3 4 3 ~B 0 0 0 0 0 0 0 0 0 0 0 0 0 75 77 84 42D 81 78 102 107 1363/14 SI S6 883 ACI RTD jA 1 1 1 1 1 ~B 1 0 0 0 0 RTD+l 000 RTD A 2 1 1 2 3 3 3 t 1 2 2 3 1 ~~~~~~B 0 0 0 0 0 0 Cultures were isolated on more than one occasion from each qf 101 different sites, i.e., quarters of cows' udders: 91 sites yielded series of 2 cultures each, 4 sites series of 3 cultures each, 3 sites series of 4 cultures each, and 3 sites series of 5 cultures each. Of these series of cultures, 44 showed no variation at all and 47 varied only in their weak reactions. In two series, completely different phage patterns were found and this was presumed to indicate infection with a different strain. These two series are not recorded below. 29 52A 3A S6 883 A 1 2 89 11 30 B 0 0 0 0 8 Collaborator 4 Seventeen cultures were tested in duplicate using, respectively, the original bovine phages from Weybridge plus the international phages and the local propagations of the bovine phages plus the international phages. Seven cultures did not vary at all and 6 varied only in their weak reactions. 80 3C 6 42E 47 54 84 42D 102 107 1363/14 S6 A l 1 1 11 1 1 1 8 4 11 9 1 B 0 0 0 2 0 0 0 0 1 3 0 0 Collaborator 5 A total of 1 034 cultures were tested twice with the phages shown below. 29 52A 3A 6 53 77 84 42D A 214 123 5 130 51 16 49 65 B 59 16 2 33 10 14 12 29 Collaborator 7 Nine cultures were retested after an 11-month interval, 4 were tested 5 times over a 6-month period, and 2 were tested 4 times over a 6-month period. Three cultures varied only in their weak reactions. 52 52A 79 80 3A 3C 55 6 42E 47 53 54 75 RTD fA 1 3 6 3 3 6 3RTD4B 1 1 4 0 0 2 0 RTD +I 000 RTD {A 1 2 3 4 4 1 5 7 11 6 6 9 6 ~B 0 0 0 2 2 0 2 2 6 0 1 2 0 91 I. DAVIDSON 77 83A 84 85 42D 81 78 102 107 1363/14 SI 883 ACI R2lD { A 1 4 1 7 1 10 R 0 B 6 7 2 2 3 0 0 2 10 2 10 2 4 7 4 5 10 11 6 8 4 3 4 6 6 2 Collaborator 10 Three direct subcultures and five single-colony subcultures from each of 10 original cultures were tested with the set B phages. Four cultures did not vary at all and the remaining 6 varied only in their weak reactions. 29 52A 42D 102 107 1364/14 Si S6 883 ACI A 3 1 3 3 3 3 1 1 2 1 Collaborator 11 Five single-colony cultures were prepared from each of 23 cultures and tested along with the original cultures. Four cultures did not vary and 6 varied only in their weak reactions. 29 52A 3A 6 53 75 77 84 42D 78 102 107 1363/14 SI ACI 7 3 1 5 2 1 1 0 4 5 3 0 1 1 5 2 3 1 2 0 8 5 15 3 1 1 3 4 2 1 Collaborator 13 Three subcultures made at weekly intervals and 3 single-colony subcultures from each of 20 cultures were tested with the set B phages. Two cultures did not vary at all and 3 varied only in their weak reactions. 29 52A 6 8 5 5 3 1 2 53 75 77 84 42D 78 102 107 1363/14 Si S6 3 0 3 1 1 4 1 8 4 1 9 0 3 5 16 3 3 2 7 1 0 Collaborator 16 311 cultures, including 65 untypable cultures, were retested at 1 000 RTD, using a slightly different technique designed to improve the detection of reactions, with certain of the phages with which they had given no reactions previously. In some of the cases where, on retesting, a reaction was found, tests were also made at RTD. 52A 79 42E 47 53 42D 81 102 1363/14 Si S6 AC1 No. of cultures retested at RTD No. + + at retest 17 4 5 0 0 0 0 3 0 1 0 3 16 4 0 0 1 2 1 6 29 6 0 29 52 52A 79 80 3A 3C 55 71 No. of cultures retested at 1 000 RTD 109 66 253 132 70 66 66 65 66 No. ++ at retest 3 0 28 16 0 0 0 0 0 6 42E 47 53 54 66 0 280 243 128 221 13 29 28 30 75 77 83A 42D 81 187 78 102 107 1363/14 S1 S6 883 ACI No. of cultures retested at 1 000 RTD No. + + at retest 68 247 153 282 151 65 69 201 273 0 3 24 4 8 0 0 27 3 262 66 66 70 141 4 2 6 0 59 A B A B 92 PHAGES FOR TYPING BOVINE STAPHYLOCOCCI Collaborator 17 In all, 154 cultures were retested with the 16 phages listed below after the cultures had been stored for 1 year. Many cultures had completely different phage patterns on retest. Reactions were both lost and gained although all the differences with phage 79 were losses. There were more losses than gains with the other group I phages also. 29 52 52A 79 80 42E 47 77 27 17 27 13 20 17 33 24 26 10 22 12 28 19 39 32 42D 81 102 107 1363/14 SI S6 AC1 39 29 10 6 60 16 55 22 36 12 26 13 82 14 34 11 Lytic spectra ofphages and phage patterns ofpropa- gating strains Tables 7 and 8 show the results obtained at Wey- bridge. The lytic spectra on the "human" test strains were determined on at least 3 separate occa- sions on different batches of medium. Those on the other test strains were determined on at least 2 occasions. The phage patterns were determined on at least 6 occasions. Ten collaborators submitted the results of their lytic spectra determinations and ten recorded the phage patterns given by the propagating strains. Only a few differences were noted between the phage patterns determined at Weybridge and those deter- mined elsewhere. The lytic spectra were more van- able, however. In particular, phages found to lyse propagating strains 6, SI, and S6 at Weybridge often gave no reactions on these strains at other labora- tories. Two collaborators compared the original seed material from Weybridge with their own propa- gations and found the two to be identical although, in each case, several of the reactions with strains 6, S1, and S6 were missing. Thus these differences appear to have been caused by differences in media and techniques rather than by changes occurring in the phages on propagation. Selection ofphages Table 9 shows the phages selected for individual collaborators and groups of collaborators by the process described under " Analysis of phage reac- tions". Where there was little to choose between very similar phages, they are shown as alternatives. Phages that were only marginally excluded by the criteria used are listed as " possibly also useful". The percentage of cultures typed at RTD with the phages definitely selected ranged from 36 to 91. In each case, these percentages are fairly close to the maximum obtainable with the available phages (see Table 4). Where a collaborator tested only a small number of cultures or where there was only a limited variety of phage patterns, it was possible to select only a few phages. Such cases, however, provide valuable evidence as to the extent to which each phage is generally useful. The following appear to be clear choices as gene- rally useful phages: group I: 29 group II: 883 group III: 6 or 75 group IV: 42D, 102, 1363/14 miscellaneous: S6 Each was chosen by the results of between 5 and 14 collaborators and by the overall results. The other phages that made a useful contribution are as follows: 52A -subdivided phage group I and I/M cultures. 3A -important for two collaborators: it signifi- cantly increased the overall percentage of cultures typed. 6 and 75 -chosen by the same number of individual results, but the overall results suggested them only as alternatives. 42E -important for two collaborators. Subdivided cultures of phage groups I, III, and IV. Signi- cantly increased the overall percentage of cultures typed. 53 and 84-each subdivided phage group III cultures and each significantly increased the overall percen- tage of cultures typed. 107 -chosen by almost as many individual results as 102 but overall overlapped more than 50% with 102 and 1363/14. Significantly increased the overall percentage of cultures typed. 78 and SI-each was chosen by several individual results. Each increased significantly the overall per- centage of cultures typed. Additional results Collaborator 5 tested a further 878 bovine cultures with the following phages: 29, 52A, 79, 3A, 6, 42E, 53, 54, 75, 84, 42D, 78, 102, 107, 1363/14, S1, ACI, 129/16, P42D/E193, 88A, 825. The results were ana- 7 A B 93 I. DAVIDSON N N N - ID] O -1 | - m Lo-£ NCICI ) CNI D t _ NO | _ ° 0 L 0~~~~~~0 N~~~~~~~~~~~~~N 0 -I 1-1 !-|- N N N I- ~I o CD o _~ C N °0_ w _ Cw X o ,CD I_ 0 00 ~ ~ ~ I~~~~~~~~~CI m LO I 0 ' N N 0 0~~~ 0 N-C I Io I, _ K lI_ N ~~~0 LO 0 r- CDtD cn uo a) N1 01 0. coWc_ co_co co2 C0D co CO CCO CC o00CD - E E~E~3 °5 o° °o Q o 0 v o CDN- -0*n U) (An _o 94 CACO 0 n C. W -o C)Ic W E .)0, .c .0 0 CO n -o a) C a) ._0 -o a) 10 c Q ._ a) C 0 1.-o 0 0 a) 0. C.). -J a) CO I- ~-I C PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 0co J J cn II I I I I I u I I .I I -~~~~~~~~~~~~~~~~~~~~~~~~l I , - I II I I IIsQn dUOS++ I+ Ii oN - -I-o+0 0 0 00+ _C) IC+ + + 0 co~~~~~~~~~~~~~~~~~~~~~o2 N + II_I+ +I_; I z 3 I+ o Io II Ii II I ; 1+l ILO 3 i+ i CYi)I II i~ I i I III I i II II I i IU - z)I 0) N1 Io 0 c/ U. 0 0 0 0 0 0 0c 0 00 0 0 0C0 0 0 0 0 N 0 r-0 w- LO 0 0 C)- N 0- N- 95 0 -C .)CD ._ .a *0 .0 CU Q ,0 D Cl) C 0) Q 0) 0. 0 0o 0 C .0 0) CO CL 0. Cu0. 0 .0 0~ o6 0 Cu H) C om 0 COo co 0 0 o 00 C14 0 CO) -0U) .C co Z Ca CO 0 E oi CD 0 OC 0 L) c - C, 0 0 aL 0cn I. DAVIDSON Table 9. Phages selected for each collaborator and for groups of collaborators Collaborator Definitely selected (including alternatives) Possibly also useful jtypeda 1 52, 42E, 75,b 42D, 107, 1363/14, S6,b 883, 367 73 2 42D, 102 or 107 or 1363/14, AC1 36 3 3A, 42D, 81, 78, 107, S6, 883 1363/14 or 42F 71 4 29, 42E, 77, 42D, 102, 1363/14, S6 80, 6 84, 78 86 5 29, 71, 75, 102,1363/14, Si 84, 42D, 107 60 6 29, 6, 42D, 102 or 107 or 1363/14, S6 86 7 6, 54, 42D, 883 80 8 29, 6, 102, 1363/14 42D, 107, S6 91 9 52A, 6, 53, 84, 1363/14 107, S1 46 10 29, 47 or 75 or 77 or 83A, 53, 42D or 102 or 107 or 1363/14, S1, 883, 3B 80b 50 11 29 or 52, 6 or 47 or 75, 77, 42D, 78, 102, 107, A13 42E, 1363/14 b 78 12 52 or 52A, 3A, 6 or 47 or 53, 54, 75, 107, 1363/14, 883 49 13 29 or 52 or 52A or 80, 75, 84, 42D, 102, 107, 1363/14, 78, S6 84 14 42D or 1363/14, 102, 107 42E,b S6 b 70 15 42D, 102 or 107 or 1363/14 70 1 6 102 or 107 or 1363/14, S1, S6 47 74 17 29, 79, 47, 102, 1363/14, S6 52 or 52A or 80, 78 42D Group 1 29 or 52, 3A, 6 or 47 or 54 or 75 or 81, 42E, 42D, 78, 107, 1363/14, Si, S6, 883 84, AC169 2 29, 52A, 3A, 6 or 47 or 54 or 75, 42E, 84, 42D, 78, 102, 1363/14, Si, S6, 883 77, 107, AC1 72 3 29, 52A, 3A, 6 or 47 or 54 or 75, 42E, 53, 84, 42D, 78, 102, 1363/14, S1, S6, 883 73 a Percentage of cultures typed at RTD with phages definitely selected. b Added after classification of cultures into lytic groups. lysed as described above and the following phages, which typed 89% of the cultures at RTD, were selec- ted: 29 or 52A, 3A, 6 or 54, 42D, 102, SI. Collaborator 6 tested a further 413 group IV cul- tures, each from a different herd, with 10 group IV phages: 42D, 42F, 102,107,108,111, 1363/14,129/16, P42D/E193, 88A. Of these, 139 cultures were sensi- tive to all the phages and 42 were untypable at RTD. Every pair of phages except those including 42D showed more than 75% similarity both ways. Phage 42D showed more than 50% similarity with the other phages. The greatest differences were shown by phages 42D, 102, and 1363/14. These three together typed 362 of the 371 typable cultures at RTD. CONCLUSION After considering the investigation reported above, the International Subcommittee on Phage Typing of Staphylococci established an international basic set of phages for typing bovine staphylococci (Int. J. system. Bact., 1971). Those phages without inter- national numbers were given the following numbers in the international series. Original no. 883 1363/14 SI S6 International no. 116 117 118 119 The international basic set for typing bovine staphylococci is thus constituted as follows: group I: 29, 52A group H: 3A, 116 group HI: 6, 42E, 53, 75, 84 group IV: 42D, 102, 107, 117 miscellaneous: 78, 118, 119 96 PHAGES FOR TYPING BOVINE STAPHYLOCOCCI 97 ACKNOWLEDGEMENTS The author thanks Dr M. T. Parker, Central Public Health Laboratory, Colindale, Middlesex, England, for his guidance and encouragement throughout this work and Mr C. J. Cousins and Mr M. Stone of the Ministry of Agri- culture, Fisheries & Food, United Kingdom of Great Britain and Northern Ireland, for writing and operating the computer programme. RIISUMT RECHERCHE COLLECTIVE DE PHAGES DESTINE~S AU TYPAGE DES STAPHYLOCOQUES BOVINS La serie de phages choisie et utilisee internationalement pour le typage des staphylocoques humains s'est r6velee peu appropriee au typage des staphylocoques bovins. I1 etait devenu necessaire de definir une batterie de phages mieux adaptee a ces demieres determinations. Sur l'ini- tiative de l'auteur, des chercheurs travaillant dans ce do- maine ont participe a une 6tude collective visant a selec- tionner des phages convenant a une utilisation g6nerale. lus ont e invites a tester au moins les phages intematio- naux 29, 52A, 3A, 6, 53, 75, 77, 84 et 42D, les phages bovins ou autres 78, 102, 107, 1363/14, SI, S6, 883 et ACI, et, dans la mesure du possible, tous les phages appartenant a la s6rie de typage des staphylocoques humains. Au total, 17 collaborateurs ont eprouve 6 999 cultures de staphylocoques bovins. On a releve de fortes variations d'un experimentateur a l'autre, dues sans doute en partie a des differences aleatoires entre les 6chantillons, surtout lorsque le nombre des cultures etait peu elev6. NMan- moins, les s6ries de r6sultats obtenus dans un meme pays ou dans des pays voisins avaient tendance A concorder davantage. Par exemple, la proportion des reactions rela- tives aux phages 3A et 883 etait elevee au Danemark et en Suede; aux Etats-Unis d'Amerique (2 laboratoires) celle des reactions relatives au phage S6; et la plus forte pro- portion des reactions avec le groupe I de phages etait observ6e en Irlande et au Royaume-Uni. L'ensemble des donn6es a et6 analyse pour evaluer l'utilite de chaque phage pour le typage des staphylo- coques bovins et pour s6lectionner un jeu de phages capa- ble de classer une forte proportion des cultures et d'assu- rer le maximum de differenciations reproductibles. Sur la base de ces resultats, le Sous-Comite internatio- nal pour la lysotypie des staphylocoques a constitu6 une serie internationale de phages pour le typage des staphylo- coques bovins. Les phages ci-dessous, depourvus a l'ori- gine d'un numero international, en ont requ un: NO d'origine NO international 883 116 1363/14 117 S1 118 S6 119 La serie internationale de base des phages a utiliser pour le typage des staphylocoques bovins s'etablit d&s lors comme suit: groupe I : 29, 52A groupe II : 3A, 116 groupe III: 6, 42E, 53, 75, 84 groupe IV: 42D, 102, 107, 117 divers: 78, 118, 119, REFERENCES Bertoni, L. & Rosaschino, F. (1963) Atti Soc. ital. Sci. vet., 17, 770 Bonin, W. & Blobel, H. (1967) Zbl. Bakt., L Abt. Orig., 205, 309 Coles, E. H. & Eisenstark, A. (1959) Amer. J. vet. Res., 20, 838 Davidson, I. (1961) Res. Vet. Sci., 2, 396 Frost, A. J. (1967) J. Hyg. (Lond.), 65, 311 Fujikura, T. & Shibata, S. (1965) Nat. Inst. Anim. Hlth. Q., Tokyo., 5,146 Gedek, W. (1966) Berl. Munch. tierdrtzl. Wschr., 79, 292 Int. Bull. bact. Nomencl., 1959, 9, 115 Int. Bull. bact. Nomencl., 1963, 13, 119 Int. J. system. Bact., 1967, 17, 113 Int. J. system Bact., 1971, 21, 167 Meyer, W. (1966) Z. med. Mikrobiol. Immunol., 152, 232 Meyer, W. (1967) J. Hyg. (Lond.), 65, 439 Mondini, S. & Dovadola, E. (1959) Zooprofilassi, 14, 755 Nakagawa, M. (1960) Jap. J. vet. Res., 8, 331 Seto, J. T. & Wilson, J. B. (1958) Amer. J. vet. Res., 19, 241 Smith, H. W. (1948) J. com. Path., 58, 179 Verge, J., Goret, P., Joubert, L., Paraf, A. & Asso, J. (1960) Rec. Mid. vet., 136, 527 Williams, R. E. 0. & Rippon, J. E. (1952) J. Hyg. (Lond.) 50, 320 I. DAVIDSON Annex COLLABORATORS 1. A. J. Frost, School of Veterinary Science, University of Queensland, Australia 2. D. Bajljosov, Central Veterinary Institute for Conta- gious and Parasitic Diseases, Sofia, Bulgaria 3. 0. Klastrup, Statens Veterinaere Serumlaboratorium, Ringsted, Denmark 4. L. Koiranen, State Veterinary Medical Institute, Helsinki, Finland 5. H. Blobel & W. Bonin, Institut fur Bakteriologie und Immunologie, Giessen, Federal Republic of Germany 6. W. Gedek, Institut fur Nahrungsmittelkunde, Miunchen, Federal Republic of Germany 7. W. Meyer, Zentrallaboratorium fur Lysotypie, Wer- nigerode, German Democratic Republic 8. J. Nyhan, The Agricultural Institute, Fermoy, Ireland 9. R. Tamarin, Kimron Veterinary Institute, Bet Dagan, Israel 10. M. Nakagawa, National Institute of Health, Tokyo, Japan 11. N. P. Markham & J. Markham, Department of Microbiology, University of Otago, New Zealand 12. H. Thorne, Veterinarbakteriologie Laboratoriet, Vasteras, Sweden 13. Ian Davidson, Central Veterinary Laboratory, Wey- bridge, and M. Patricia Jevons, Central Public Health Laboratory, Colindale, United Kingdom 14. G. P. Shamanova & M. V. Kirikova, Dairy Institute, Vologda, USSR 15. N. P. Nefedjeva, Institute of Nutrition, Moscow, USSR 16. E. R. Garrison, T. E. Patrick & L. T. Patterson, Department of Animal Sciences, University of Arkansas, USA 17. L. W. Slanetz, Department of Microbiology, Uni- versity of New Hampshire, USA 98