
 
        
         
		Marine,  or  brackish water:  frequent.  Poole  Bay,  June  1848;  Seaford,  
 Sussex,  May  1850;  Bexhill,  Sussex, Oct.  1852;  Belfast  Bay,  July  1853;  
 Cork  Harbour,  April  1855,  &c.,  IF. Sm.  Kinghorn,  Fife,  Nov.  1853, D/.  
 GreviUe.  NevjhieDeposit,Prof.Harkness.  Var./3. Itford,Sussex,March 1853;  
 Newhaven,  Sussex, Nov. 1853.  Var. y.  Seaford, Sussex, April  1854, TF. Sm. 
 The  synonymy  of  this  geiras  is  much  confused,  aud  without  authentic  
 specimens, it is impossible to  determine species from  the description or figures  
 of authors.  I  believe  that  all  the  above  forms are to be referred to  a single  
 species,  which is pretty  constant  in every locality iu wliich it has  occurred to  
 me.  The  only differences  I  have  noticed,  consist  in  the  greater  or less development  
 of the  connecting  mucus-cushion,  and  the presence or absence of  
 the  curious keel-like rim of silex :—forms  aberrant in these respects are, however, 
  so frequently intermixed witli the ordinary frustules, that I cannot regard  
 ■  such peculiarities as of specific importance. 
 All filamentous species of Diatomaceæ being probably stipitate on their first  
 production, the free or attached condition of the Melosireæ cannot be regarded  
 , as even  constituting varieties,  aud  the M.  salina  ft.  of  Kützing  must  therefore  
 be discarded. 
 Plate XLIX.  329. 
 2.  Melosira  Borrerii,  Grev.  Ordinary  and  sporangial  frustules  
 subcylindrical,  all  geminate ;  polar  usually much  shorter than  the  
 equatorial  diameter ;  valves  subhemispherical,  distinctly  cellulate ;  
 cingulum marked with  conspicuous  circles  of  cellules.  Breadth  of  
 ordinary  filament -OOll" to -0022".  Breadth  of sporangial  filament  
 •0031" to  -0052".  v.v. 
 Var.  ft.  Filaments very much curled. 
 Grev  B .P .p   401.  Ealfs, Ann. vol.  12.pl.ix. 2. ad specim. authen.  in herb. 
 T  P'-  ’"■•  Gdllionella  lineata, Ehr. 
 Int  X .  i  0 ?  Melosira moniliformis, Kùtz. Bacill. iii. 2. ad specim.  quæ  
 dedit  cl.  De  Brebisson.  Melosira  nummuloides, Ag.  Consp.  p.  65  ad  
 specim.  authen.  in Herb.  Grev. 
 Marine,_or brackish water.  Poole Bay,  in great abundance, Nov.  1848 and 
 i   i  raas  n  n  z.  dY;  “ "n?-  W. Sm.  Aberdeen, 
 ra w  lSi)4, DI r  . AAr ?ioFtt.F   V4 a'  r.  Nea‘ r Lewes,  Feb.  1852,  ,S'wz. Clyde,  July 
 This  species  is  readily  distinguished  from  the  last  by  its  larger  size,  and  
 the globular and more  cylindrical form  of its frustules.  Its cellulate structure,  
 which  gives  a  rich  brown hue to  the desiccated filament,  also affords  an easy  
 means of identification.  It would  appear,  from  a  note  in the  ‘ Conspectus ’  
 that Agardh regarded this  species  as identical with hh Melosira nummuloUe's-,  
 tor  he  says,  “ Specimina  ex  Shoreham  Harbour  mihi a Borrero  communi-  
 cata  omnino  couveniunt  cum  specimiiiibus,  quæ  ipse  cum  Iloffinaimo  Icgi ;” 
 and  yet  he  gives  as  synonyms  Conferva  nummuloides,  Eng.  Bot.  tab.  2287,  
 and Fragilaria nummuloides, Lyng.  Tent.  tab.  63,  both of which  species  are  
 referred  by  most  authors  to  our  first  species,  and  the  figures of  which  are  
 certainly more accordant with that  than with our M. Borrerii.  The words in  
 the Conspectus,  “ articulis circularibus,”  are also inapplicable  to  the specimens  
 from  Shoreham,  and  are  correctly  descriptive  of  our M. nummuloides.  It  
 is probable,  therefore,  that Agardh was  unable with  his  instrument to  discriminate  
 between these forms,  and assigned them  in  different  states  to  different  
 species.  In  uncertainty,  I  adopt the  name  of M. Borrerii,  under which,  in  
 honour  of  an  excellent  native  observer,  the  species  was  first  described  in  
 Britain,  and  whose  discovery  of  it  in  Shoreham  Harbour  appears  to  have  
 been almost simultaneous with the determination of Agardh’s M. nummuloides,  
 and venture  to  hope  that  the  description  and  figures  now given will prevent  
 future  misunderstanding.  The  sporangia  of M. Borrerii  are  of  great  size,  
 and when in this  state its filaments run no risk of being confounded with  any  
 of its allies. 
 Plate L. 330. 
 3.  Melosira  subflexilis,  Kiitz.  Frustules  at  first  cylindrical,  frequently  
 united by  a mucus-cushion, geminate ;  mature  and  sporangial  
 frustules  subglobular ;  young valves  cylindrical,  with  a  slight  
 constriction  between  the  apex  and  suture ;  mature  valves  sub-  
 hemispherical ;  extremities of valves always convex.  Breadth of filament  
 •0002"  to  •0008".  Breadth  of sporangial  filament-0011".  v.v. 
 Kütz. Bacill. ii.  13.  Raben.  Süssw. Diat.  ii.  Conferva  lineata, Dillw. Conf.  
 tab. B.  Lyng. Tent.  Ixiii. B. 
 Fresh or brackish water.  'IVareham,  Dorset,  Nov.  1849 ;  Lewes-,  Sussex,  
 December 1851, March  1852 ;  Denton, Sussex, November 1853, W.Sm.  Carrickfergus, 
   September  1853 ;  Largs,  July 1854, Dr. Arnott.  Aberdeenshire,  
 June 1848, Dr. Dickie. 
 Plate LI. 331. 
 4.  Melosira  varians,  Ag.  Ordinary  frustules  cylindrical;  sporangial  
 at  first  globular,  at  length  cylindrical;  valves  cylindrical,  
 with  truncated  extremities.  Breadth of  filament  -0002"  to  •OOll".  
 Breadth  of sporangial  filament  •0015".  v.v. 
 Ag.  Consp.  p.  64.  ad  specim.  authen.  in  Herb.  Grev.  Ralfs, Ann. vol.  12.  
 pl. ix. 5. ad  specim. in Herb.  Jenn.  Kûtz. Bacill. ii.  10. ad specim. quæ  
 misit am. De  Brebisson.  Hass.  Alg.  xciii.  4 &  5.  Thw.  Ann.  2nd ser.  
 vol. 1. pl. xi. A.  Prit. Anim. xxiv. 32.  GalUonella varians, Ehr. Inf. x. 4 .  
 Conferva fasciata, Dillw. Conf.  tab.  B. 
 Fresh water :  very frequent.  Wiiiterbourne,  Lewes,  April  1852.  Itford,