SUMMARIES OF PLANT CHARACTERS, ETC. 





that are the same as one or the other parent or both 

 parents the differences are small. Among the macro- 

 scopic characters, 85.3 per cent are inter) liate, and 



there is a very small combined percentage of characters 

 developed in excess and deficit of parental extremes 

 (5.9 per cent). Among the microscopic characters 

 49.4 per cent are intermediate and 42.5 per cent are 

 developed beyond parental extremes. Summing up the 

 percentages of characters that are intermediate and that 

 are developed beyond parental extremes, respectively, 

 it is seen that of the total characters 60 per cent are 

 intermediate and 32.4 per cent developed beyond 

 parental extremes. 



In the companion hybrid, Cypripedium latkamianum 

 inversum (Chart FT, Table I, 1), the macroscopic and 

 microscopic characters are found to be closely in accord 

 in their percentages with those of the C. latkamianum, 

 the most noticeable differences being in the percentages 

 that fall under the characters that are the same as the 

 pollen parent and those that are intermediate. In this 

 hybrid the percentage of macroscopic characters that 

 are the same as those of the pollen parent is larger than 

 the percentage of microscopic characters; but in the 

 other hybrid the reverse. The percentages of both macro- 

 scopic and microscopic intermediate characters are less, 

 especially as regards the former. In this hybrid 73.5 

 per cent and in the other 85.3 per cent of the macro- 

 scopic characters are intermediate, while the figures for 

 the microscopic characters are 4(5.6 and 49.4, respec- 

 tively. Summing up the characters that are interme- 

 diate and those that are developed beyond parental ex- 

 tremes, respectively, it is seen that of the total characters 

 54.1 per cent are intermediate and 36.5 per cent de- 

 veloped beyond parental extremes. This gives in this 

 hybrid in comparison with the other a lower percentage 

 of characters that are intermediate and a larger percent- 

 age that are developed in excess and deficit of parental 

 extremes. The corresponding percentages and hence 

 the corresponding curves of these hybrids are so closely 

 alike that one should at a glance suspect that the plants 

 are very closely related. In fact, the similarities and 

 dissimilarities noted are generally in accord with what 

 should naturally be expected from the data of hybrids. 



The remarkable degree of concordance of the data 

 of these two hybrids is a matter of pre-eminent impor 

 tance because of the data of one being in the nature of 

 a check-off or test experiment in relation to the other. 

 It is obvious if the data do not agree within limits thai 

 have been found by the systematist in his descriptions 

 of the naked-eye characters of plants, that they would be 

 regarded as being (independable, and that if, on the 

 other hand, they do agree that the differences in the 

 corresponding percentages in the macroscopic and micro 

 scopic characters are not fallacious. It scarcely seems 

 within the realm of possibility, if the data were not 

 reliable within reasonable or small limits of error of 

 observation, that the two sets of curves would be so 

 Dearly alike and differ only to about the degree that 

 should be expected in the case of offspring of reciprocal 

 crosses. There is also, as will be seen, a distincl likeness 

 of the courses of the curves of the chart of Cypripedium 

 nitens to those of the preceding < 'ypript dium charts, and 

 the differences between the former and the latter are defi- 



niti !• more marked, thus indicating that the parentage 



in the two The ' in be 



accounted for in pari by the fa< t thai one of I 



of C. nitens (C, t Mot am) is also a parent < 



the other hybrids the pollen parent in the lir.-t and 



- d parent in I ad. The chart 



and ( '. lat) n inversum are more alike I 



of C. nitens and C. latkamianum; in both of the formi r 

 the seei! parent is the same; and. as will b< 

 out later in sufficient detail, C. villosum is more ; 

 in influencing the characters of the hybrids than is either 

 of the other parents, which in a measure will account for 

 similarities of all three chart-. 



In Cypripedium nil, ns (Chai t I I. 1) the 



percentages of both m i pic and micro arac- 



ters thai are the as ed parent and 



that are developed in exi ess of pan otal extremi s are dis- 

 tinctly larger, and there ar- lowerings of per- 

 centages of both macroscopic and mi interme- 

 diate charai ters. There is a more marked difference be- 

 tween the i 1 E macroscopic characters that are 

 the same as those of both parents, with, moreover, an 

 inversion of the ma and microscopic values in 

 this phase; and the ma lie and miero-eopic per- 

 centages of characters that are developed in 

 parental extremes are practically the same, whereas in 

 the other two hybrids they are very different. The 

 macroscopic percentages arc higher than the microscopic 

 percentages among the character-; that are the sate 

 those of the seed parent and that are intermediate, but 

 lower in the other four sex-phases. Of the total num- 

 ber of macroscopic characters 50 per cent are interme- 

 diate and 34.4 per cent are developed in excess and 

 deficit of parental extremes; and of the microscopic 

 characters 35 per cent are intermediate and 47 per cent 

 are developed in exci bs or deficit of parental extremes. 

 Summing both macroscopic and microscopic characters, 

 39 per cent are intermediate and 42.4 per cent are de- 

 veloped beyond parental extremes. The corresponding 

 figures for C. latkamianum are 60 and 32.4, and for 

 C. latkamianum inversum 54.1 and 36.5, showing in 

 C. nitens an inversion of these sex-phase value- com- 

 pared with the values of the other two hybrids. 



By comparing Charts F 1 to F S it will be seen that 

 while there arc throughout certain well-defined n 

 blances, no two are so similar, even in the case of the 

 two Cypripedium hybrids that have come from recipro- 

 cal crosses, as to lead to one being mistaken for an- 

 other. A common plan of distribution of percei I 

 of characters among the six parent-phases is evident 

 in all of the charts and is only exceptionally departed 

 from — that is. in general, comparatively low pi 

 of characters that are ihe same as one or the other 

 parent or both parents, generally higher pi 

 of characters that are developed in excess or deficit of 

 parental extremes, and still higher percentages of 



that arc intermediate. Departures of modifi- 

 cations of this plan are seen particularly in Ipoi 

 sloteri, in the higher percentage of characters developed 

 in excess of parental extremes than of intermediate eliar- 



; and in Witt wna in the high p 



of macros that are the same as those 



of the seed and pollen parent. Perhaps there is no: 



